1541
User's
Guide
1541
User's
Guide
by
Gerald Meufeld, Ph.D.
Brandon University
Technical Illustrations by
Diane M. Corralejo
HDATAMOSr
19821 Nordhoff Street, Northridge, CA 91324
(818) 709-1202
Brady
ISBN 0-89303-738-9
Copyright • 1984 by DATAMOST, Inc.
All Rights Reserved
This book is published and copyrighted by DATAMOST, Inc. All rights are reserved
by DATAMOST, Inc. Copying, duplicating, selling or otherwise distributing this
product is hereby expressly forbidden except by prior written consent of
DATAMOST, Inc.
The words COMMODORE, CBM, COMMODORE 64, VIC-20, and the Commodore
logo are registered trademarks of Commodore Business Machines, Inc.
Commodore Business Machines was not in any way involved in the writing or other
preparation of this book, nor were the facts presented here checked by them for
accuracy.
The information presented in this book is the result of intensive study and testing of
the operation of the 1541 and DOS 2.6. Every effort has been made to provide error
free information. However, neither the author nor DATAMOST, Inc. can accept
responsibility for any loss or damage, tangible or intangible, resulting from use or
misuse of this information.
Printed in U.S.A.
Acknowledgements
I would like to acknowledge the patience and forbearance of my family and friends.
Without their support, producing this manual would have been considerably more
difficult.
I would like to extend special thanks to Dr. Richard Immers for getting me started on
writing this book and to Dr. Tom MacNeill and my wife, Nancy, for their diligent
work in proofreading the manuscript. Special thanks also go to Jim Butterfield and
Mike Todd for reading the manuscript and suggesting worthwhile improvements.
This manual was written on a Commodore computer system using the WordPro 4
Plus™ word processing system available from Professional Software Inc. This excel-
lent word processor made the project much easier.
TABLE OF CONTENTS
How to Use This Book 11
Chapter 1 — Getting Started 15
1.1 Setting Up Your System 15
1.2 Powering ON Your System 16
1.3 Inserting a Diskette 17
1.4 Using a 1541 with a VIC-20 18
1.5 Rules for Using Floppy Diskettes 18
1.6 Viewing a Diskette's Directory 19
1.7 Wild Cards and Pattern Matching 21
1.8 Creating Selective Directory Listings 22
1.9 Loading Programs 24
1.10 Listing Programs and Directories on a Printer 28
1.11 Preparing a New Diskette for Use 29
1.12 Saving Programs 30
1.13 Verifying Programs 33
1.14 Where Next? 35
Chapter 2 — Using DOS 5.1 37
2.1 Loading DOS 5.1 37
2.2 Displaying a Directory 38
2.3 Loading a Program 39
2.4 Loading and Running a Program 41
2.5 Saving a Program 41
2.6 Sending a Disk Command 42
2.7 Reading the Error Status 42
2.8 DOS 5.1 Commands in a Program 43
2.9 Tips on Using DOS 5.1 Effectively 43
Chapter 3 — Using the Command Channel 47
3.1 The Command Channel 47
3.2 Opening the Command Channel 47
3.3 Sending Disk Commands 48
3.4 Combining OPEN and PRINT# Statements 49
3.5 Reading the Command Channel 50
3.6 Reading the Error Status in a Program 51
3.7 Reading the Error Status in Immediate Mode 52
3.8 Closing the Command Channel 53
Chapter 4 — Diskette Housekeeping 57
4.1 Formatting a New Diskette (full NEW) 58
4.2 Reusing a Diskette (short NEW) 61
4.3 Initializing a Diskette 64
4.4 Scratching a File 66
4.5 Renaming a File 68
4.6 Copying a File ...,,. 71
4.7 Combining Two or More Files 73
4.8 Validating a Diskette 76
4.9 Your BASIC Housekeeper ! 79
Chapter 5 — Introduction to File Handling 89
5.1 Separating Data from a Program 89
5.2 What a Data File Is 90
5.3 Kinds of Files 91
5.4 Accessing a File 92
5.5 Opening a File 93
5.6 The Logical File Number 94
5.7 The Device Number 95
5.8 The Channel Number 96
5.9 File Identification Information 97
5.10 Disk Drive Status 101
5.11 Limitation on the Number of Disk Files Open 103
5.12 Writing to a File 103
5.13 Monitoring the Number of Blocks Free 105
5.14 Reading from a File 106
5.15 Detecting the End of a File 109
5.16 Closing a File 110
5.17 Conclusion Ill
Chapter 6 — Program Files 113
6.1 Introduction to Program Files 113
6.2 Structure of a Program File 114
6.3 Loading a Program File 115
6.4 Saving a Program File 118
6.5 Verifying a Program File 121
6.6 Replacing a Program File 123
6.7 Inside a BASIC Program 124
6.8 Using a Program File 129
6.9 The Directory as a Program File 1 34
6.10 Self- modifying Programs 1 37
6.1 1 Typical Application (mail list) 140
Chapter 7 — Sequential Files 147
7.1 Introduction to Sequential Files 147
7.2 Structure of a Sequential File 148
7.3 Using a Sequential File 148
7.4 Replacing a Sequential File 156
7.5 Appending Data to a Sequential File 158
7.6 Handling Very Large Sequential Files 160
7.7 Storing a Program as a Sequential File 163
7.8 Typical Application (mail list) 165
Chapter 8 — Relative Files 173
8.1 Introduction to Relative Files 173
8.2 Structure of a Relative File 174
8.3 Planning a Record 174
8.4 Deciding on the Record Length 176
8.5 Using a Relative File 176
8.6 The Problem with the Position Command and a Fix 186
8.7 Replacing an Existing Relative File 191
8.8 Adding Data to a Relative File 192
8.9 Typical Application (mail list) 192
Chapter 9 — Indexed Relative Files 201
9.1 Introduction to Indexed Relative Files 201
9.2 Organization of the Files 202
9.3 Planning an Application of Indexed REL Files 203
9.4 Keeping an Index 204
9.5 The Binary Chop 205
9.6 Programming the Binary Chop 208
9.7 Alternate Uses for the Binary Chop 209
9.8 Programming Considerations for Indexed REL Files 210
9.9 Typical Application (mail list) 212
Chapter 10 — File Handling from Machine Language 225
10.1 Introduction to Machine Language File Handling 225
10.2 Opening a File .... 225
10.3 Reading a File 228
10.4 Writing a File 229
10.5 Closing a File 231
10.6 Monitoring the I/O Status 232
10.7 Monitoring the Disk Drive Status 234
10.8 Sending Disk Commands 235
10.9 Typical Application (machine language file copy) 236
Chapter 11 — Getting Out of Trouble 247
11.1 Spotting and Diagnosing Problems 247
11.2 Unscratching a File 252
11.3 Recovering Data from an Unclosed File 254
11.4 Recovering from a Short NEW 260
1 1.5 Repairing a Damaged Directory 260
11.6 Recovering from a Full NEW 266
11.7 Recovering a Damaged File 267
11.8 Recovering from a Bad Replacement 269
1 1 .9 Recovering a Physically Damaged Diskette 272
1 1.10 Conclusion 273
Chapter 12 — Care and Maintenance 275
12.1 General Operating Hints 275
12.2 Cleaning the Record/Play Head 276
12.3 Permanently Changing the Device Number 276
12.4 The Alignment Problem 278
12.5 Checking the Alignment 279
12.6 Realigning Your 1541 285
12.7 Preventing Drive Wheel Slippage 287
Chapter 13 — A Bit of Background 291
13.1 Inside a Floppy Diskette 291
13.2 What Diskettes to Buy 292
13.3 The Back Side Controversy 293
13.4 The Recording Process 294
13.5 The Sync Mark 296
13.6 How a Sector is Organized 297
1 3.7 How the Directory is Organized 299
13.8 Modifying the Directory 307
1 3.9 How Forward Pointers Work 313
1 3.10 How Sequential Files are Stored 318
13.11 How Program Files are Stored 319
13.12 How User Files are Stored 320
13.13 How Relative Files are Stored 320
1 3.14 An Overview of the Inner Workings of DOS 2.6 328
13.15 The Important Bugs in DOS 2.6 330
Appendix A — Glossary 333
Appendix B — Reference Tables 343
Appendix C — Disk Error Messages 347
Appendix D — Summary of Disk Commands 353
Appendix E — Disk Utility Program Listings 361
Index 411
Application Programs:
Program Name
MAIL PGM
MAILSEQ
MAIL REL
MAIL INX
FAST COPY
Disk Utility Programs:
Purpose
A self-modifying mail list
A mail list using SEQ files
A mail list using REL files
An advanced mail list using
indexed relative files
A fast, machine language
file copier program
Page
142
167
194
214
237
Program Name
HOUSE HELP
RECOVER UNCLOSED
ALLOCATE
FIND ERRORS
REDO 18/0
TRACE CHAIN
RECOVER FILE
ALIGN 1541
MOD ENTRY
TRACE FILE
FULL DIRECTORY
CONFIRM ALL FILE
BACKUP
MAKE FAST COPY
EDIT T&S
MOD DISK NAME
FIX DIRECTORY
UNSCRATCHER
PRG HEX DUMP
PRG ANALYZER
Purpose
Makes diskette housekeeping easy
Recover data from unclosed file
Allocate blocks in unclosed file
Scan track 18 for errors
Recover from errors on 18/0
Display chain given T&S pointer
For files not in directory
Check alignment or realign
Modify name, type, or length
Display chain given file name
Expanded directory listing
Check that all files are OK
Make a backup of a diskette
Create a fast file copy program
Edit any byte on a diskette
Modify a diskette's name
Repair a damaged directory
Unscratch any disk file
Hex display of BASIC program
Byte by byte look at a program
Page
79
257
258
261
262
268
270
282
309
317
361
364
369
374
376
386
387
400
403
406
10
How to Use This Book
This book is intended to help you make more effective use of your Commodore disk
drive. Although the examples and program listings are designed for use on a
Commodore 64 or VIC-20 with a 1541 disk drive, many will work without modification
on other Commodore computers and disk drives.
This book expands and clarifies the documentation provided by Commodore Business
Machines, Inc. Some of the information covered in the 1541 User's Manual is also
presented here. However, the majority of information presented in this book is original
and is the result of intensive research and testing by the author.
Beginners
If you have just purchased a 1541 disk drive, you should read the first two chapters
carefully. The first chapter explains how to set up and operate your disk drive. The
second chapter explains how to make use of the DOS 5.1 program included on the
1541TEST/DEMO diskette that came with your disk drive. These two chapters
should get you started. If you have trouble understanding computer jargon, just
check the glossary in Appendix A.
As your collection of diskettes grows you will reach the point where you will need to
learn how to get rid of unwanted programs, erase and reuse an old diskette, etc.
These tasks are often called diskette housekeeping. Chapter 4 tells you how to do
these housekeeping tasks and explains what happens to your diskette as these tasks
are performed. This chapter concludes with a listing of a menu-driven BASIC
program that makes housekeeping easier. If you make a mistake, check Chapter 11
to see if you can rectify the situation.
If you are the type of person who likes to know how things work, you will want to
read Chapter 13 carefully. It will help to take some of the mystery out of the inner
workings of your disk drive.
Intermediate Programmers
If you already know how to operate your disk drive and know how to write simple
BASIC programs, you may want to skim through Chapters 1 and 2 quickly. Once you
understand how to use the command channel to read the disk drive status and do
diskette housekeeping, you are ready for file handling. Make certain that you
understand the fundamentals of file handling (Chapter 5) and how to use the various
file types (Chapters 6 to 8) before you tackle the chapters on indexed relative files or
machine language file handling.
11
Experienced Programmers
As an experienced programmer, you probably have some specific application in
mind. Use the Table of Contents and the Index to locate the information or techniques
you need. However, don't ignore the more introductory sections. They contain a great
deal of information that has never been published before. For example, did you know
that: DOS 5.1 has a save command, DOS 5.1 commands can be used within a
program, sequential files can be saved and loaded, or that there is a Modify (M) mode
that allows you to read an unclosed file?
Note that the chapter on file handling from machine language uses only the standard
Commodore 64 Kernal jump table. If you use other ROM entry points, your program
may not work on some future versions of the Commodore 64 because there is no
guarantee that Commodore will retain the same entry points.
About the Programs
This book contains many listings of ready to use programs written in BASIC. There
are three main types of programs: instructional, application and utility.
Instructional Programs
Many chapters contain one or more programs that are designed to help you learn
more about how your computer or disk drive works. Some of these programs will not
have much lasting value. Others, such as the one to merge programs or those to
analyze a program, will probably find a place in your tool kit.
Application Programs
Each of the main chapters on file handling from BASIC concludes with a mailing list
program. Chapter 10 on machine language file handling concludes with a fast file
copy program. These programs are designed to illustrate a practical application of
the file handling techniques discussed in that chapter. Even if you do not bother to
type in these programs, you should study them to be certain that you understand the
techniques used. Although the programs work quite well as is, you may want to
modify them by adding color or sound. Some possible modifications and improve-
ments are suggested at the end of the program listings.
Utility Programs
This book also contains several disk utility programs. Some of these are scattered
throughout the text. Several more are given in Appendix E. These are powerful
tools. Remember, any tool can be dangerous. Be sure you read and understand the
operating procedures before you use one of these programs. It is always a good idea to
practice using a disk utility program with a test diskette before you use it with one
that contains valuable programs or data.
Typing the Programs
Program listings in books and magazines often suffer from two problems: typo-
graphical errors that occur when the program is retyped into a word processor, and
the readability of the Commodore's control characters (e.g., the reverse-field heart
that means clear the screen). To overcome these problems, the program listings for
12
this book were created using a special lister program. The lister program takes a
working program and converts it into a WordPro file. At the same time, the control
characters are converted into English language code words surrounded by curly
brackets. For example, the reverse-field heart which is used to clear the screen is
converted to {CLR}. The table below lists the code words, their meaning, and the
proper key or keys to press on your Commodore 64 or VIC-20.
Code
Meaning
Press these Keys
{CLR}
Clear the screen
.Hold SHIFT and
press CLR/HOME
{HOME}
Home the cursor
Press CLR/HOME
{DOWN}
Cursor down
Press CRSR
(up/down)
{UP}
Cursor up
Hold SHIFT and
press CRSR (up/down)
{RIGHT}
Cursor right
Press CRSR
(right/left)
{LEFT}
Cursor left
Hold SHIFT and
press CRSR
(right/left)
{RVS}
Reverse field ON
Hold CTRL and
press 9
{ROFF}
Reverse field OFF
Hold CTRL and
press 0
Notes:
1. When a number appears inside the curly brackets, it means you repeat the oper-
ation immediately to the left of the number, that many times. For example, {DOWN
5} means to press CRSR (up/down) five (5) times.
2. All programs have been listed in a column 40 characters wide. Except where
special characters have been spelled out between curly brackets, the lines are
listed exactly as they appear on a Commodore 64 display. Spaces must be typed in
as listed. Where necessary, count the character columns to determine the appro-
priate number of spaces.
Using the Programs
This book contains listings of over fifty programs written in BASIC. These programs
are copyrighted. They may not be used commercially nor may they be sold. Since
many of the programs are long, typing them all in would be a time consuming,
tedious process. Feel free to share your typing efforts with your friends who have also
purchased a copy of this book. However, please do not share these programs with
those who do not own a copy of the book.
If you do not have the time to type in the programs, you may purchase a diskette
containing all the programs from DATAMOST.
13
CHAPTER
1
GETTING STARTED
This chapter is designed for the new 1541 owner. It covers such topics as setting up
your system, operating procedures, running a program and saving a program.
Experienced 1541 users may want to skim through this chapter quickly.
1.1 Setting Up Your System
Before you unpack your new drive you should decide where you are going to put it. It
should be on a sturdy, level surface near your computer. If your computer is on a desk
or table, you can set the drive beside the computer or on a shelf above or below the
desk or table top. Try to keep it at least six inches away from your monitor or TV set
(stray magnetic fields can cause problems). Avoid locations where you will be tempted
to set a cup of coffee or ashtray on it.
Remember that you will need a place to store your diskettes. They should go on a shelf
or in a drawer. You do not want to leave them lying around loose. Be sure you have
lots of space, diskette collections tend to grow rapidly!
When you unpack your 1541 you should find the following items in the packing case:
1. The 1541 disk drive.
2. A grey three-prong power cord.
3. A black serial bus cable with a DIN connector on each end.
4. A copy of the 1541 User's Manual.
5. A copy of the 1541TEST/DEMO diskette.
If one or more pieces are missing, contact the store where you bought it. It is a good
idea to keep the box and protective packaging in case you move or have to take your
drive in for service.
You may see a cardboard tab sticking out of your 1541. Pull on it gently and remove
the floppy diskette-sized piece of cardboard. Keep it. You should slip it into your drive
whenever you will be transporting your 1541. The cardboard helps protect the
read/write head.
Follow these steps to set up your disk drive:
Step 1: Take the female end of the grey power cable and plug it into the back of the
disk drive. Plug the other end into any grounded, three-prong electrical outlet.
15
WARNING:
If the drive makes any sounds when you plug it in, turn OFF the disk drive before you
go any further. The power switch is at the back of the drive. Be sure both your
computer and disk drive are OFF before you connect any other cables.
Step 2: Plug the black serial bus cable into either of the serial bus sockets on the back
of your disk drive. It doesn't matter which end of the cable you use, they're identical.
Left: Rear view of 1541 disk drive.
Below: Rear view of the Commodore 64
computer.
Step 3: Make sure your computer is switched OFF. Now, plug the other end of the
black serial bus cable into the back of the computer. If you do not have a printer or
other peripherals to attach, the hookup is now complete.
Step 4: If you have a printer or a second disk drive, plug its serial bus cable into the
second serial port on the back of the 1541.
1.2 Powering ON Your System
Once you have connected all the parts of your computer system, it is time to start
switching it ON. Your 1541 User's Manual suggests that your computer should be the
last thing you switch on. However, many people who had several peripherals attached
to the serial bus encountered difficulties. As a result, Commodore released a technical
bulletin suggesting the following power ON sequence.
16
Computer -> disk drive
Computer -> disk drive -» printer
Computer -» disk drive -» disk drive
Computer -> disk drive -> disk drive -> printer
Note that the computer is always switched on first and the printer, if any, is switched
on last. The use of a power bar to switch everything ON at once is not recommended;
particularly if you have several peripherals hooked up to the serial bus. The system
may lock up.
WARNING:
Always remove all diskettes from a disk drive before you turn it OFF or ON. If a
diskette is left in the drive, you may create a magnetic spike on it that may make
important data or programs unreadable.
When you switch the disk drive ON, the motor will start, the green power-on LED
will come on, and the red drive-active LED will light up. After a few seconds, the
motor will stop and the red drive-active LED will go out. It is now safe to insert a
diskette and begin working.
1.3 Inserting a Diskette
To open the door of your 1541 drive, simply press the door catch lightly inwards. The
catch will move upward and the door will pop open. Any diskette in the drive will be
ejected automatically when the door is opened. To insert a diskette, slide it into the
drive with the label side up. The edge nearest the large opening goes in first. The
write protect notch should be on the left like this:
^7=^
w
Write
Protect ^
Notch
A
17
The diskette should slide in easily. If it sticks or binds, slowly pull it out and try again.
When the diskette is almost completely into the drive, you will feel a slight spring
resistance. This is normal. You will know that you have inserted the diskette com-
pletely when you hear a click and the spring resistance stops. Close the disk drive
door by pushing downward on the catch. When the catch has moved downward about
half an inch it will spring outward. The door is now closed and the diskette is firmly
locked into place. You may now begin working with the diskette.
WARNINGS:
1. Always remove the diskette before you switch your disk drive OFF or ON.
2. Never open the drive door while the red drive-active LED is glowing steadily and
the motor is running. If you do this while the drive is writing, part of your
program or data may be lost. For the one time when you should ignore this rule,
see Section 11.6.
1.4 Using a 1541 with a V1C-20
The 1541 User's Manual suggests that, if you are using a 1541 with a VIC-20, you
should send a "UI-" command to the 1541 to switch it to VIC-20 data transmission
speed. Current 1541 drives do not require this speed adjustment. As a result, this
section of the 1541 User's Manual may be ignored.
1.5 Rules for Using Floppy Diskettes
Floppy diskettes are quite reliable but they must be handled with care. Follow these
rules to ensure long and satisfactory service from your diskettes:
1. Keep things clean. Dust and grit will cause rapid wear of your diskettes and the
1541's record/play head. Don't set up your system in a dusty environment. It is a
good idea not to smoke while using your system.
2. Do not turn the disk drive ON or OFF while a diskette is in the disk drive. If you
do, the surge of current may produce a magnetic spike on the diskette surface.
Depending on where this spike occurs, you may lose valuable data or programs.
3. Never touch a diskette's recording surface; handle it by its protective jacket. If you
touch the recording surface, the oil from your fingers will contaminate the surface
and make the diskette unusable. Be particularly careful about the back side of the
diskette because the data is stored on that side.
4. Keep your diskettes cool and dry. Diskettes work best at normal room temperature.
Too high a temperature can cause the disk jacket to warp and make the diskette
unusable. Too cold a temperature can cause moisture problems. If you must trans-
port your diskettes and disk drive in very cold weather, wrap them in a plastic
bag and allow them to warm up in the plastic bag for at least an hour before using
them. This prevents moisture from condensing on them.
5. Keep your diskettes away from magnetic fields. Electric motors, TV sets and
other electrical appliances produce magnetic fields that can destroy the programs
and data recorded on your diskettes. This means you should never put your
diskettes on top of your monitor!
18
6. Put your floppy diskettes back into their paper storage envelopes when you are not
using them. This helps to protect them from dust.
7. Make backup copies of important programs and data files. You never know when
disaster may strike!
1.6 Viewing a Diskette's Directory
One floppy diskette can store many different programs. In order to make use of a
program stored on a floppy diskette you need to know its name. The names of all the
programs and data files stored on a diskette are kept in the diskette's directory. To
view a diskette's directory on the Commodore 64 or VIC-20 you must LOAD it into
memory and then LIST it.
Let's practice this using the 1541TEST/DEMO diskette that comes with your 1541.
Here is what to do:
Step 1: Insert the 1541TEST/DEMO diskette into your disk drive.
Step 2: Load the directory by typing the following command and pressing the
RETURN key.
LOAD "*",8 or LOAD "*0",8
When you enter this command the red drive-active LED will light up and the motor
will run. After a few seconds the drive-active LED and motor should turn OFF. If the
drive-active LED keeps flashing after the motor turns off, a disk error has occurred.
Turn your computer OFF and ON and try again. If you cannot load the directory of
the 1541TEST/DEMO diskette, take your drive back to where you bought it.
HINT:
Instead of turning your Commodore 64 OFF and ON to reset things, you can hold
down the RUN/STOP key and press the RESTORE key. The screen should clear and
the READY prompt and the cursor should appear. Now type in SYS 64738 and press
the RETURN key. Presto!
Step 3: List the directory by typing LIST and pressing the RETURN key. You
should see the following display on your video screen.
0 "1541 TEST/DEMO " ZX 2A
13 "HOW TO USE" PRG
5 "HOW PART TWO" PRO
4 "VIC-20 WEDGE" PRG
1 "C-64 WEDGE" PRG
4 "DOS 5.1" PRG
11 "COPY/ ALL" PRG
9 "PRINTER TEST" PRG
4 "DISK ADDR CHANGE" PRG
4 "DIR" PRG
6 "VIEW BAM" PRG
19
4 "CHECK DISK" PRG
14 "DISPLAY T&S" PRG
9 "PERFORMANCE TEST" PRG
5 "SEQUENTIAL FILE" PRG
13 "RANDOM FILE" PRG
558 BLOCKS FREE.
Let's look at this directory listing piece by piece.
O "1541 TEST/DEMO " ZX 2A Drive number
The first line begins with the drive number zero. This is a carry over from the
Commodore dual drives. It will always be zero on your 1541.
O "1541 TEST /DEMO " ZX 2A Diskette name
The drive number is followed by the diskette name. This is the name specified when
the diskette was prepared for use (see Section 1.11 on formatting a diskette). The
name is padded with blank spaces to make it exactly 16 characters long.
0 "1541 TEST/DEMO " ZX 2A Diskette formatting ID
The next two characters are the diskette identification (ID) characters. They were
also specified when the diskette was prepared for use. In this case the diskette ID is
ZX.
O "1541TEST/DEM0 " ZX |2A Diskette DOS type
The last two characters on the first line identify the diskette format. These will be 2A
on any diskette formatted for use on a 1541 drive.
&£B BLOCKS FREE. Storage space remaining
The last line of the directory listing indicates how much space is still available on the
diskette for storing new programs or data. In this case there are 558 blocks of storage
space left. Each block can hold 256 characters (bytes) of information. On a newly
formatted diskette there are 664 blocks available for your use (free). This means that
you can store a total of 664 x 256 = 169,984 characters (bytes) of information.
The lines between the first and last lines are the directory entries for the various
programs and data files stored on the diskette.
13 "HOW TO USE" PRG
5 "HOW PART TWO" PRG
4 "VIC-20 WEDGE" PRG
1 "C-64 WEDGE" PRG
4 "DOS 5.1" PRG
20
11 "COPY/ ALL" PRS
9 "PRINTER TEST" PRG
4 "DISK ADDR CHANGE" PRG
4 "DIR" PRG
6 "VIEW BAM" PRG
4 "CHECK DISK" ! PRG
14 "DISPLAY TScS" PRG
9 "PERFORMANCE TEST" PRG
5 "SEQUENTIAL FILE]' PRG
13 "RANDOM FILE" I PRG
This diskette contains a total of 15 directory entries. There is one entry for each file of
information stored on the diskette; The listing gives the size, name and type of each of
these files. Let's examine the first entry in detail.
S3 "HOW TO USE" PRG Length of file
The number on the left indicates the size (length) of the file in blocks. The file
corresponding to this entry is 13 blocks long.
13 "HOw " USE" PRG Filename
The directory entry also gives the name of the file. This file is named HOW TO USE.
You must know a file's name in order to make use of the file. A file name may be up to
16 characters in length. Almost any combination of letters, numbers or graphics
symbols may be used for a file name.
13 "HOW TO USE" !f|||| File type
The last part of the directory entry is the file type. All the files on the 1541TEST/
DEMO diskette are program (PRG) files. The other common file types are:
File type Entry
Sequential file
SEQ
Relative file
REL
User file
USR
Now that you know how to load and list a diskette's directory to find out what files are
stored on the diskette, let's see how wild cards can make your life a bit easier.
1.7 Wild Cards and Pattern Matching
In order to make use of a program or data file stored on a diskette you must give its
name. On a Commodore disk drive you do not have to give a file's complete name
when you want to access it. You can use wild card characters that will match one or
more characters in the name. There are two wild card characters: the question mark
(?), and the asterisk (*). The question mark (?) will match any single character in a
name. The asterisk (*) will match any number of characters at the end of a file name.
21
When using wild cards you type some combination of ordinary characters (A, B, C,...)
and wild card characters (? and *). This creates a pattern. When you send the pattern
to the drive, it will compare the names of the programs and data files in the directory
with the pattern and see which ones match. This is called pattern matching.
Let's take a look at a few examples of patterns to see how this works. Well begin with
the question mark (?). Since the ? matches any single character in a file name, the
pattern ?EST will match the names BEST, TEST and REST. However, since it
matches one and only one character, it will not match BLEST, GUEST or TEST1.
The ?EST pattern specifies a four character name with any first character followed
by the letters EST.
The * will match any number and combination of characters at the end of a name.
This makes it more powerful. For example, the pattern CHE* will match any name
starting with the three letters CHE such as CHECK DISK, CHECKERS or
CHESTNUT. If you use the * in a pattern, there is no point in putting any characters
following the *, such as TE*ER, because any ending will match.
You may use both wild cards in a pattern. For example, the pattern ?ON* will match
any name having ON as the second and third characters in the name (e.g., BONZO,
CONSUMER GRIPES and CONTEST).
You can use wild cards whenever you have to type the name of a program or data file
stored on a diskette. You will find them handy for loading programs, doing disk
housekeeping, and accessing files. They are especially handy for creating selective
directory listings.
1.8 Creating Selective Directory Listings
Occasionally you will run across a diskette with many, many files stored on it (the
directory can hold up to 144 entries). In this case you may want to look at only some of
the entries. All you have to do is include a file name pattern in your LOAD command
like this:
SYNTAX:
LOAD "*:xxxxM,8
ALTERNATE:
LOAD ,,*0:xxxx,,,8
EXAMPLES:
LOAD "*:DIR",8
LOAD "*:?IR",8
LOAD "*:DI*",8
LOAD "*0:?I*",8
WHERE:
xxxx = a file name pattern which may contain one or more wild card
characters.
22
If the load command contains a complete file name with no wild card characters, the
directory listing will contain the diskette name line, the one entry whose name you
specified, and the blocks free line like this:
Command: LOAD "*:DIR",8
Resulting O "1541 TEST/ DEMO " ZX 2A
listing: 4 "DIR" PRO
558 BLOCKS FREE.
If there is no file on the diskette with the name you specified, you will just get the first
and last lines of the directory display.
To create a selective listing containing more than one entry you use the wild card
characters (? and *) to create a pattern that will match more than one file name. Let's
examine the ? first.
Since the ? will match any single character in a file name, the pattern ?IR will match
the names VIR, DIR and SIR. However, because it matches one and only one
character, it will not match WHIR, HAIR or NAIR.
Command: LOAD ,,*:?IR",S
Resulting 0 "1541 TEST /DEMO " ZX 2A
listing: 4 "DIR" PRG
558 BLOCKS FREE.
The * matches any number and combination of letters at the end of a file name. For
example, the pattern DI* will match any name starting with the two letters DL
Command: LOAD "*:DI*",8
Resulting O "1541 TEST /DEMO " ZX 2A
listing: 4 "DISK ADDR CHANGE" PRB
4 "DIR" PRG
14 "DISPLAY T&S" PRG
558 BLOCKS FREE.
The two wild cards can be used together. The * is always used last because it will
match any number or combination of letters from that point on. For example, the
pattern ?E* will match any file with an E as the second letter in its name.
Command: LOAD "*:?E*",8
Resulting 0 "1541 TEST/DEMO " ZX 2A
listing: 9 "PERFORMANCE TEST" PRG
5 "SEQUENTIAL FILE" PRG
558 BLOCKS FREE.
Practice using pattern matching to load selective directory listings until you get a feel
for how to use the wild card characters. It will be useful later for loading programs
and doing diskette housekeeping.
23
1.9 Loading Programs
Now that you know how to examine a diskette's directory you can start to make use of
the programs stored on the diskette. In order to use a program that is stored on a
diskette you must bring a copy of the program into the computer's memory. This is
called loading the program. Loading a copy of a program does not destroy or affect
the copy of the program stored on the diskette. To bring a copy of a program into
memory use a LOAD command like this:
SYNTAX :
LOAD "name", 8 (Program relocated)
ALTERNATE:
LOAD name*, 8
SYNTAX:
LOAD " n ante " , 8 , 1 (Program not relocated)
ALTERNATE:
LOAD name*, 8, 1
EXAMPLES:
LOAD "HOW TO USE", 8
LOAD "HOW T*".8
LOAD "DOS 5.1 ",8,1
F* = "VIEW BAM"
LOAD F*,8
WHERE:
name = the name of the desired program. It may contain one or
more wild card characters.
8 = the device number of the disk drive.
1 = the channel number (secondary address).
name$ = the name of a string variable containing the desired
program name.
NOTE:
Our last example illustrates how you can store part or all of the file name in a string
variable. If you are not familiar with string variables, don't worry. Just type out the
file name.
If several different program names match the one specified in the load command, the
first program in the directory that matches the specified name will be loaded. For
example, the name DI* matches three different entries in the directory: DISK ADDR
CHANGE, DIR, and DISPLAY T&S. The command LOAD "DI*",8 will load the
first entry, DISK ADDR CHANGE.
As soon as you press the RETURN key, the drive motor will start and the red
drive-active LED will light up. The program is being loaded. After a few seconds, or
minutes if the program is a very long one, the drive motor should stop, the red
24
drive-active LED should go out, your computer should print the word READY on the
screen, and the flashing cursor should return. This indicates that the program has
been loaded successfully. If the computer prints the message, ?FILE NOT FOUND
ERROR, the program name you gave in the LOAD command does not match the
name of any program stored on that diskette. Check the directory and try again.
If the cursor doesn't return and your computer appears to have locked up, hold down
the RUN/STOP key and press the RESTORE key. This should get the cursor back.
The program did not load properly. Check the directory and try again.
Once the program has been loaded into the computer's memory simply type the word
RUN and press the RETURN key to begin the program. Some commercial programs
have an auto-run feature. These programs will start running automatically as soon as
they have been loaded into memory. An auto-run program must be loaded using the
LOAD "name",8,l form of the LOAD command. Most auto-run programs cannot be
listed.
Practice loading and running a program using the program named HOW TO USE
on your 1541TEST/DEMO diskette. This program gives a brief description of the
various programs on the diskette. Here is what you should see on your screen as you
do this:
LOAD "HOW TO USE", 8 Remember to press RETURN.
SEARCHING FOR HOW TO USE Computer response.
LOAD I NG Computer response.
READY. Load completed successfully.
Once a program has been loaded into the computer's memory you can list the program
by typing LIST and pressing the RETURN key. To slow down the listing, hold down
the CTRL key. To stop the listing, press the RUN/STOP key. Most commercial
programs are protected and cannot be listed. If you see something like this when you
list a program, 0 SYS 2038, the program is either a machine language program or a
compiled BASIC program and cannot be listed.
HINTS:
1. Instead of pressing RETURN after typing a LOAD command, try this. Type a
colon (e.g., LOAD "COP*",8:) and then hold down the SHIFT key while you press
the RUN/STOP key. The program will not only load, it will run automatically.
2. The 1541TEST/DEMO diskette contains a program that tests your disk drive to
make sure that it is working properly. The program is called PERFORMANCE
TEST. You should load and RUN this program to check out your new drive. You
will need either a new, blank diskette or one that you don't mind being erased to
use this program.
You may be wondering about the significance of the ",1" at the end of some of the
LOAD commands. The ",1" is a signal to your computer to load the program into the
same part of memory from which it was saved. This is used mainly when loading
25
commercial programs or special machine language routines. It is not needed when
loading a BASIC program. There is only one non-BASIC program on your
1541TEST/DEMO diskette, DOS 5.1. Try loading this program without the M" in
the load command like this:
LOAD "DOS 5.1", 8
When this program has finished loading, list it by typing the word LIST and pressing
the RETURN key. You should see something like this:
52940 NEXTNEXTNEXTNEXTNEXTFORFORFORNEXTN
EXT!!!!! GGG 7.//~~ >>@#Q
43690
++++++++
DOS MANAGER V5. 1/071382
BY BOB FAIRBAIRN
<C> 1982 COMMODORE BUSINESS MACHINES
41676 EXPWAIT
7SYNTAX ERROR
READY.
This is definitely not a normal BASIC program. It is a very useful machine language
program that belongs in an entirely different part of memory. To load it into its
proper location you would have to give the command:
LOAD "DOS 5.1 ",8,1
However, there is a program that does this automatically for you. It is called C-64
WEDGE. Load this program normally and list it Here is what you should see:
LOAD "C-64 WEDGE", 8
SEARCHING FOR C-64 WEDGE
LOADING
READY.
LIST
10 IFA=OTHENA=l: LOAD "DOS 5.1", 8,1
20 I FA=1THENSYS 12*4096+ 12*256
30 NEW
READY.
The C-64 WEDGE program is what is known as a loader program. Line 10 loads the
machine language program DOS 5.1 into its proper place in memory (note the ",1").
Line 20 activates the routine with a SYS command (begin running a machine
26
language routine that starts at the location specified). Line 30 wipes out the loader
program. Many commercial programs make use of loader programs to load program
segments, sprites and high resolution pictures into memory. Usually the loader
program also checks to make sure that you are using a legitimate copy of the program
and not a bootleg copy. As a result, loaders are often designed so that their mode of
operation is obscure.
You can use a directory listing to make loading a program even easier. Put your
1541TEST/DEMO diskette in your drive and follow these steps:
Step 1: Load and list the directory like this:
LOAD
"*"!,8
SEARCHING FOR *
LOADING
READY.
LIST
0 "1541 TEST/DEMO " ZX
2A
13
"HOW TO USE"
PRG
5
"HOW PART TWO"
PRG
4
"VIC-20 WEDGE"
PRG
1
"C-64 WEDGE"
PRG
4
"DOS 5.1"
PRG
11
"COPY/ ALL"
PRG
9
"PRINTER TEST"
PRG
4
"DISK ADDR CHANGE"
PRG
4
"DIR"
PRG
6
"VIEW BAM"
PRG
4
"CHECK DISK"
PRG
14
"DISPLAY T&S"
PRG
9
"PERFORMANCE TEST"
PRG
5
"SEQUENTIAL FILE"
PRG
13
"RANDOM FILE"
PRG
558 BLOCKS FREE.
READY.
Step 2: Move the cursor up so that it is beside the name of the program that you want
to load and type the word LOAD over the blocks column. Then use the cursor right
key to move the cursor to the end of the name and type ",8:". For example, if you
wanted to load COPY/ALL, your display would look like this:
0 "1541 TEST/ DEMO " ZX 2A
13 "HOW TO USE" PRG
5 "HOW PART TWO" PRG
4 "VIC-20 WEDGE" PRG
1 "C-64 WEDGE" PRG
4 "DOS 5. 1 " PRG
27
LOAD "COPY / ALL", 8: _ PRO Cursor on this line.
9 "PRINTER TEST" PRG
4 "DISK ADDR CHANGE" PRG
4 "DIR" PRG
6 "VIEW BAM" PRG
4 "CHECK DISK" PRG
14 "DISPLAY T&S" PRG
9 "PERFORMANCE TEST" PRG
5 "SEQUENTIAL FILE" PRG
13 "RANDOM FILE" PRG
558 BLOCKS FREE.
READY.
Step 3: Don't worry about the file type, PRG, behind the name. To load the program
into memory, just press the RETURN key. The messages, SEARCHING FOR,
LOADING and READY, will overwrite the listing, but this doesn't matter. If you
used the load only option, be sure to clear the screen once the load is completed.
Otherwise you may press RETURN on one of the other lines and enter junk into
your program.
0 "1541TEST/DEM0 " ZX 2A
13 "HOW TO USE" PRG
5 "HOW PART TWO" PRG
4 "VIC-20 WEDGE" PRG
1 "C-64 WEDGE" PRG
4 "DOS 5.1" PRG
LOAD "COPY/ ALL", 8: PRG
9 "PRINTER TEST" PRG
SEARCHING FOR COPY ALL" PRG
LOADINGIR" PRG
READY. VIEW BAM" PRG
"CHECK DISK" PRG
14 "DISPLAY T&S" PRG
9 "PERFORMANCE TEST" PRG
5 "SEQUENTIAL FILE" PRG
13 "RANDOM FILE" PRG
558 BLOCKS FREE.
READY.
Now that you know how to load and run a program, let's take a look at how you can
make a permanent copy of a program or a diskette's directory.
1.10 Listing Programs and Directories on a Printer
Normally, once a copy of a program or diskette directory has been loaded into memory
you can list it. Simply type LIST and press the RETURN key. If you are not familiar
with examining program listings on your computer, here are a few pointers:
LIST Lists the entire program or directory.
LIST 100-150 Lists only lines 100 to 150 inclusive.
LIST -200 Lists all lines up to line number 200.
LIST 125- Lists line 125 and all subsequent lines.
28
Press RUN/STOP to abort the listing.
Press CTRL to slow down the listing.
If you have a Commodore compatible printer, you can make a printed copy of the
listing by entering the following commands:
OPEN 3,4 Open file to printer.
CMD3 Direct output to printer.
LIST (or some variation) Output begins.
Once the printer stops, enter the following commands:
PRINT# 3 Switch output back to screen.
CL0SE3 Close file to printer.
HINTS:
1. If you are using a Commodore printer and would like your program listings
double-spaced instead of single-spaced, simply use any number between 128 and
255 in place of the three in the commands listed above.
2. If you forget to enter the PRINT# and CLOSE commands, the cursor on the
SuF ™may n0t behave Pr(>Perly. It may stay at the end of a line when you press
the RETURN key. If this happens to you, you can force the cursor onto a new line
by holding down the SHIFT key when you press RETURN. To get the cursor to
respond properly again, you will either have to power down, or enter CLOSE3 and
then repeat the OPEN, CMD, PRINT#, CLOSE sequence (you don't have to
LIST).
1.11 Preparing a New Diskette for Use
NOTE:
Before you buy any new, blank diskettes read Section 13.2. It gives you some idea of
what to look for.
A new, blank diskette has nothing recorded on it. Before it can be used to store
programs or data files, a pattern of tracks and sectors must be recorded onto the
surface using your 1541 disk drive. This is called formatting or newing the diskette.
Note that the NEW command that we will be using for this job is a disk command. It
is no* the same thing as the BASIC command you give to get rid of the program in
memory. To format a blank diskette follow these four steps:
Step 1: Turn ON your computer system. Remember that your Commodore 64 or
VIC-20 is switched on first (see Section 1.2).
Step 2: Insert the blank diskette into the disk drive.
Step 3: Start the formatting process by entering a NEW command into your
computer like this. Remember to press the RETURN key at the end of the line.
29
syntax:
OPEN 15,8, 15,"N:disk name, id"
EXAMPLES:
OPEN 15,8, 15, "N: MY TEST DISKETTE, Tl"
OPEN 15,8,15, "N: GREAT GAMES, Gl"
OPEN 15,8,15,"N:DATA BASE #1,DB"
OPEN 15,8, 15, "N: INCOME TAX, 85"
WHERE:
disk name = the diskette name that will appear at the top of any
directory listing. The name may be up to 16 characters
long. You can use letters, numbers, spaces and graphics
symbols. You may not use an asterisk (*), question mark
(?), comma (,) or colon (:).
id = a two character diskette identification code. Use a different
combination of characters for each of your diskettes.
CAUTION:
Use a different ID for each diskette you format! It is a good idea to develop a systematic
numbering system for your diskettes as soon as you buy your first box of diskettes.
This will save a lot of renaming, renumbering and recopying later. For example, you
might number your word processing diskettes Wl, W2, . . . , your games diskettes Gl,
G2 etc.
As soon as you press the RETURN key to enter the command, the red drive-active
LED will light up and the drive motor will turn on. There will then be a clattering
noise for a second or so. This is normal. After about 80 seconds the drive-active light
and the drive motor should turn off. If the drive-active light continues to flash after
the drive motor stops, the formatting operation was not successful. Close the file (Step
4) and try again. If the diskette will not format correctly after several tries, it is
defective— get rid of it!
Step 4: Close the command channel by entering the following command into your
computer. Remember to press the RETURN key at the end of the line.
SYNTAX:
CLOSE 15
EXAMPLE:
CLOSE 15
More information about the disk command NEW and the formatting process is given
in Section 4.1.
1.12 Saving Programs
When you take the time to type in a program, you will probably want to record a copy
of it on a diskette so you won't have to type it in again. This is called saving the
program. Saving a copy of a program onto a diskette does not destroy or affect the
30
copy of the program in the computer's memory. To save a copy of a program in
memory you use a SAVE command. It is similar to a LOAD command. It looks like
this:
SYNTAX : Remember to press
SAVE "name", 8 the RETURN key after
typing in a command.
ALTERNATE:
SAVE name*, 8
EXAMPLES:
SAVE "HOW TO USE", 8
SAVE "HOW PART TWO", 8
F* - "MY PROGRAM"
SAVE F*,8
WHERE:
name = the name that will appear in the directory.
name$ = the name of a string variable containing the program name
desired.
8 = the device number of the disk drive.
Choosing a Name
You may save your program using any name you like. The disk drive does not care
whether the name you select has any relationship to the nature of the program. But,
to save you endless hours of frustration, choose a name you can remember. Otherwise,'
as your collection of diskettes grows, you will be unable to find anything.
A program name may be up to 16 characters long. It may not contain an asterisk (*), a
question mark (?) or a comma (,). In addition, the name may not begin with an "at"
sign (@). A name may contain one or more graphics characters, shifted spaces, reverse
field characters, or null characters, CHR$(0). However, the use of these characters is
discouraged. To use any special characters in a name, you will have to store the name
as a string variable.
As soon as you press the RETURN key, the drive motor will start and the red
drive-active LED will light up. The program is being saved. After a few seconds, or
minutes if the program is a very long one, the drive motor should stop, the red
drive-active LED should go out, your computer should print the word READY on the
screen, and the flashing cursor should return. This indicates that the program has
been saved successfully.
If the cursor doesn't return and your computer appears to have locked up, hold down
the RUN/STOP key and press the RESTORE key. This should get the cursor back.
The program did not save properly. However, the program is still in the computer's
memory. Change diskettes and try again.
31
NOTES:
1. A new diskette must be formatted first. Before you can save any programs onto a
new, blank diskette, it must be formatted. If you forgot to do this earlier, just
follow Steps 2 to 4 in Section 1.11. You will not lose your program in memory.
2. Watch the drive's drive-active light. If the drive-active LED is flashing after the
drive motor stops, the program has not been saved properly. Don't panic, your
program is still in the computer's memory. Simply insert a different diskette and
try again. The most common causes of an error are: the diskette has not been
formatted, the diskette is full or the write protect notch is covered by a tab. See
Sections 3.5 and 5.10 for more information about the error status. messages.
Sections 2.7, 3.6 and 3.7 tell you how to read these error messages. Appendix C
contains a complete list of the error messages and what they mean.
3. You can only save normal programs this way. The simple SAVE command given
above can only be used to save a copy of a BASIC or machine language program
that resides in the part of memory normally occupied by a BASIC program
(starting at 2049 on a Commodore 64, 4097 on an unexpanded VIC-20, or 4609 on a
VIC-20 with an 8/16K expander). To save a different part of memory see Section
6.4.
Practice saving a program by typing in the short demo program listed below and
saving it. Here is what your screen should look like as you do this:
NEW Dumps old program.
READY. Computer response.
10 REM TEST PROGRAM Enter test program.
20 INPUT"HI, WHAT'S YOUR NAME" 5 N*
30 PRINT" IT WORKS, ";N*
SAVE " TESTER " , 8 Save a copy on disk.
SAVING TESTER Computer response.
READY. Computer response.
Once the program has been saved, you may want to VERIFY it to be sure that it has
been saved correctly (see Section 1.13). Now, load a copy of the diskette's directory
and list it. Your screen should look something like this:
LOAD " * " , 8 Load the directory.
SEARCHING FOR * Computer response.
LOAD I NG Computer response.
READY. Computer response.
LIST
0 "MY TEST DISKETTE" Tl 2A
1 "TESTER" PRG
663 BLOCKS FREE.
READY.
32
Finally, load the program back in again and run it. This time your screen should look
like this:
LOAD " TESTER " , 8 Load a copy from disk.
SEARCHING FOR TESTER Computer response.
LOADING Computer response.
READY. Computer response.
RUN You type in RUN.
HI, WHAT'S YOUR NAME? GERRY You type in your name.
IT WORKS, GERRY Computer response.
READY. Computer response.
As a final check to make sure that you know how to save a program, load a copy of
HOW TO USE from the 1541TEST/DEMO diskette into memory and save a copy
onto another diskette. The command sequences for loading and saving programs may
seem a bit complicated now, but, with a little practice, they will become second
nature.
1.13 Verifying Programs
After you have saved or loaded a program, you may want to check that the copy of the
program stored on the diskette is an exact copy of the program in memory. This is
called verifying the program. Verifying a copy of a program does not destroy or
affect the copy of the program in the computer's memory or the one stored on the
diskette. It merely compares the two copies. To verify a copy of a program in memory
use a VERIFY command like this:
SYNTAX:
VERIFY "name", 8 (Relocated program)
ALTERNATE:
VERIFY name*, 8
SYNTAX:
VERIFY "name", 8,1 (Non-relocated program)
ALTERNATE:
VERIFY name*, 8,1
EXAMPLES:
VERIFY "HOW TO USE", 8
VERIFY "HOW PART TWO", 8
VERIFY "DOS 5.1 ",8,1
F* = "MY PROGRAM"
VERIFY F*,8
33
WHERE:
name
8
1
name$
the name of the program as it appears in the listing of the
diskette's directory (wild cards may be used).
the device number of the disk drive.
the channel number (secondary address).
the name of a string variable containing the program's
name.
As soon as you press the RETURN key, the drive motor will start and the red
drive-active LED will light up. The program is being verified. After a few seconds, or
minutes if the program is a very long one, the drive motor will stop, the red drive-
active LED should go out, and your computer will print one of two messages , OK or
VERIFY ERROR, followed by the word READY and the flashing cursor. The
message OK indicates that the two copies of the program match exactly. The message
VERIFY ERROR indicates that the two copies do not match exactly.
NOTES:
1. The copies must match exactly. Unless the two copies match exactly, the VERIFY
ERROR message will appear. Even one extra space will do it!
2. A relocated program will not verify correctly. When you load a BASIC program
into your Commodore 64 or VIC-20 that was saved onto a diskette using a different
type of Commodore computer (PET, VIC or Commodore 64), the program is
automatically relocated in memory so that it will run properly. A program that
has been relocated will not verify correctly because the line links have been altered.
See Chapter 6 for more information about the line links and the relocation process.
3. You can verify machine language routines. If you use the ",1" form of the VERIFY
command, you can verify a copy of any portion of the computer's memory.
Practice verifying programs by typing in this short demo program, saving it and
then verifying it. Here is what your screen should look like as you type in, save and
then verify a copy of the short demo program.
NEW
READY.
10 REM TEST PROGRAM
20 INPUT "HI, WHAT'S YOUR NAME";N*
30 PR I NT "GOOD WORK, ";N*
SAVE "NEW TESTER", 8
SAVING NEW TESTER
READY.
VERIFY "NEW TESTER", 8
SEARCHING FOR NEW TESTER
VERIFYING
OK
READY.
Dumps old program.
Computer response.
Enter test program.
Save a copy on disk.
Computer response.
Computer response.
Verify the copy on disk.
Computer response.
Computer response.
Copies match response.
Computer response.
34
Now add a single space to line 10 of the program so that it looks like this:
10 REM TEST PROGRAM Two spaces after TEST.
Now verify the program again. Your screen should look like this:
VERIFY-NEW TESTER", 8 Verify the copy on disk.
SEARCHING FOR NEW TESTER Computer response.
VER I F Y I NG Computer response.
? VER I FY ERROR Copies don't match response.
READY. Computer response.
QUESTION:
Should I always verify a program after I save it? In general, the 1541 disk drive is a
pretty reliable device. Whenever information is recorded onto a diskette, the drive
reads it back automatically to make sure that it has been recorded correctly. As long
as you use good quality floppy diskettes, keep your drive in good working order, and
use good work habits, you should never have to VERIFY a file. It will always be
saved correctly. However, for your own peace of mind, it is always a good idea to
verify important files and keep a backup copy in case anything happens. Appendix E
contains two programs to help you. Use the BACKUP program to make a duplicate
copy of an entire diskette. The FAST COPY program makes a duplicate copy of a
single program or data file.
1.14 Where Next?
The information contained in this chapter should get you well started in using your
1541 disk drive. The next chapter describes how you can use the DOS 5.1 program to
simplify the basic disk operations such as loading or saving programs. If you are
interested in how your disk drive works, take a look at Chapter 13. As your diskette
collection grows you will need to consult Chapter 4 on diskette housekeeping. If you
run into trouble, check Chapter 11. Once you have mastered the basics, you will
probably want to learn about file handling. For this, Chapter 5 is a good place to start
35
Chapter
2
USING DOS 5.1
Your 1541TEST/DEMO diskette includes a powerful utility, DOS 5.1, that makes
using your disk drive easier. DOS 5.1 is a short, machine language program that
resides at the top of your computer's memory and adds several, disk related commands
to BASIC. It is designed so that it does not affect the operation of normal BASIC
programs. However, it may not be compatible with some commercial, machine
language programs.
The best way to learn about DOS 5.1 is by trying the various commands yourself. To
get the most benefit out of this chapter you should try all the exercises on your
computer. So grab your 1541TEST/DEMO diskette and get your computer fired up.
2.1 Loading DOS 5.1
Now that you're all set, insert your 1541TEST/DEMO diskette into your 1541 and
let's get started. The first thing to do is load DOS 5.1 and run it. Your 1541TEST/
DEMO diskette contains two versions of the DOS 5.1 program. One is for use with the
Commodore 64 and the other for the VIC-20. Do not load DOS 5.1 directly. Use a
loader program. Use the appropriate command to load the loader program:
Commodore 64 VIC-20
LOAD "C-64 WEDGE", 8 LOAD "VIC-20 WEDGE", 8
The loader program is called a wedge because the DOS 5.1 program is going to
wedge (hook) into BASIC to add the new, disk related commands. Once the WEDGE
program has loaded, RUN it. After a few seconds you should see a display something
like this:
DOS MANAGER V5. 1/071382
BY BOB FAIRBAIRN
(C) 1982 COMMODORE BUSINESS MACHINES
READY.
37
This display indicates that the DOS 5.1 program is active. Once you see the DOS 5.1
screen display, you are ready to begin.
2.2 Displaying a Directory
Let's begin our exploration of the DOS 5.1 commands by displaying a diskette's
directory. With your 1541TEST/DEMO diskette in the disk drive, type in either of
the following commands:
>* or e*
NOTES:
1. The two characters > and @ take on a special meaning when the DOS 5.1 program
is active. They become command characters. Whenever your computer encounters
either of these characters as part of a line you have entered, the DOS 5.1 program
checks the characters to the right of the command character to see if this is a DOS
5.1 command. If it is, the command is carried out. The two command characters
are interchangeable. Take your pick.
2. Although the > is now a command character, you can still use it as a normal
"greater than" symbol in a BASIC program.
As soon as you press the RETURN key, the drive-active LED should turn on and the
drive motor should start. In a few seconds you should see the following screen display:
0 "1541 TEST/DEMO " ZX 2A
13 "HOW TO USE" PRG
5 "HOW PART TWO" PRO
4 "VIC-20 WEDGE" PRG
1 "C-64 WEDGE" PRG
4 "DOS 5.1" PRG
11 "COPY/ ALL" PRG
9 "PRINTER TEST" PRG
4 "DISK ADDR CHANGE" PRG
4 "DIR" PRG
6 "VIEW BAM" PRG
4 "CHECK DISK" PRG
14 "DISPLAY TStS" PRG
9 "PERFORMANCE TEST" PRG
5 "SEQUENTIAL FILE" PRG
13 "RANDOM FILE" PRG
558 BLOCKS FREE.
This is the same display that you would get if you loaded the directory and listed it.
However, there is one important difference. The program in memory is not erased
when you display the directory using DOS 5.1. Let's type in a test program and see
how this works. Here's what to type.
38
NEW Type NEW and press RETURN to erase
any program in memory.
READY. Computer response.
10 FOR K=l TO 10
20 PRINT "IT WORKS!"
30 NEXT
Now display the diskette's directory again by typing:
>* or @*
Once the directory has been displayed, type LIST and press the RETURN key. Your
program is still there!
Have you ever listed a long directory and had the name you wanted scroll off the
screen? With DOS 5.1 you can pause a directory listing by pressing the SPACE BAR.
You can resume the listing by pressing the SPACE BAR again. You don't really need
to use this feature with the 1541TEST/DEMO diskette because the directory is short
enough to fit on the screen. However, practice using the SPACE BAR to pause or
resume the listing so you feel comfortable with this feature.
With DOS 5.1 you can even use the wild cards (? and *) to display a selective directory
listing. Let's try some examples to see how this works:
Command Resulting directory listing
>*:DI* O "1541 TEST/DEMO " ZX 2A
4 "DISK ADDR CHANGE" PRG
Any name that 4 "DIR" PRO
starts with DI 14 "DISPLAY T&S" PRG
558 BLOCKS FREE.
>*:D?S* O "1541 TEST /DEMO " ZX 2A
4 "DOS 5.1" PRG
Any name starting 4 "DISK ADDR CHANGE" PRG
with D and with S 14 "DISPLAY T&S" PRG
as the third letter 558 BLOCKS FREE.
>*:??E* 0 "1541 TEST /DEMO " ZX 2A
6 "VIEW BAM" PRG
Any name having E 4 "CHECK DISK" PRG
as the third letter 558 BLOCKS FREE.
Now that you know how to display a diskette's directory using DOS 5.1, let's take a
look at how to load a program.
2.3 Loading a Program
You can use two different commands, / and %, to load a program. The first, /, is used
for loading BASIC programs. The second, %, is used for loading machine language
routines, character sets, hi-res graphics screens, etc.
39
Let's try the / command first. Well use it to load in the program called HOW TO
USE. Here's what your screen should look like:
/HOW TO USE The load command.
SEARCHING FOR HOW TO USE Computer response.
LOADING
READY.
Note that you did not have to type quotation marks around the program's name. You
can put the name in quotes if you want to (e.g., /"HOW TO USE"), but you are not
required to have the name in quotes.
The reason that the / command is used for loading BASIC programs is that it relocates
the incoming program. This means that the incoming program will be stored in your
computer's memory in the proper location for a BASIC program. You can load any
BASIC program that was written on a Commodore computer into your computer's
memory. The program may not run properly if it uses a lot of PEEKs or POKEs, but
at least it will load properly.
Now let's try the % command. We'll use it to load the same program again. Here's
what your screen should look like:
"/.HOW TO USE The load command.
SEARCHING FOR HOW TO USE Computer response.
LOADING
Oops, what's that garbage doing all over the screen?
If you are using a Commodore 64, your screen is now covered with junk. What
happened? The program HOW TO USE was not written on a Commodore 64, it was
written on a Commodore PET. When you used the % command to load it, the program
was not relocated. It loaded into the same part of memory that it originally occupied
in the Commodore PET. Unfortunately, this area of the Commodore 64's memory is
the screen.
Because the % command does not relocate the program, it is used mainly for loading
machine language routines, hi-res screen images, sprite images, etc.
SPECIAL NOTE:
At this point you must either turn your computer OFF and ON and reload DOS 5.1 or
enter one of the following commands:
POKE 2048, 0 Commodore 64
POKE 4096,0 VIC-20 (unexpanded)
POKE 4608,0 VIC-20 (8 or 16K expander)
none needed VIC-20 (3K expander)
40
If you forget to do this, you will get a ?SYNTAX ERROR whenever you attempt to
run a program. The reason is that the memory location immediately below the start
of BASIC must contain a zero. The zero that is normally there was changed when we
loaded the program into the wrong part of memory. The POKE command restores
the normal value.
Using Wild Cards in File Names with DOS 5.1
You can use the wild cards (? and *) as part of the program name in either type of
load command. For example, try loading the program VIEW BAM using the
command:
/VI*
If several different program names match the one specified in the load command, the
first program in the directory that matches the specified name will be loaded. For
example, the name DP matches three different entries in the directory, DISK ADDR
CHANGE, DIR and DISPLAY T&S. The command, /DP, will load the first entry,
DISK ADDR CHANGE.
2.4 Loading and Running a Program
Thet command is used to load and run a BASIC program. The program is relocated
as it is loaded. There is no DOS 5.1 command to load and run a program without
relocating it. Use the following command to load and run the program HOW TO
USE from the 1541TEST/DEMO diskette.
f HOW TO USE
As with the / and % commands, you can use the wild cards (? and *) as part of the file
name. The following command will load and run the program VIEW BAM:
f VIE*
Now that you know how to use DOS 5.1 to load, or load and run programs, let's take a
look at how it makes it easier to save a program.
2.5 Saving a BASIC Program
The <- command is used to save a copy of the BASIC program currently in memory.
To experiment with this command you will need to use a formatted diskette that does
not have a write protect tab on it. If you are using a brand new, blank diskette, be
sure to format it first (see Section 1.11). Let's begin by typing in this short test
program:
10 FOR K=l TO 10
20 PRINT "THIS IS A TEST"
30 NEXT
Now let's save the program onto diskette using this save command:
<-TEST #1
41
As soon as you press the RETURN key, the drive-active LED should light up and the
drive motor should turn on. After a few seconds, the light should go out and the motor
turn off. The current disk error status (see Section 2.7) will be displayed to the right
of the SAVE "filename" message. If the message is "00, OK,00,00," the program has
been saved correctly. You can double check by using the >$ command to display the
directory to be sure that it has been saved properly.
2.6 Sending Disk Commands
Disk commands are messages sent to the disk drive to tell the Disk Operating System
(DOS) to carry out a particular operation. There are commands to prepare a diskette
for use, rename a file, scratch (eliminate) a file name from the directory and so on.
Since these disk commands are usually used for eliminating unwanted files or re-
arranging the files in a diskette collection, they are often called housekeeping
commands.
DOS 5.1 makes it very easy to send these diskette housekeeping commands to the
1541 disk drive. All you do is type either a > or @ command character followed by the
disk command you want. For example, either of these commands will erase the
program, TEST #1, from our diskette:
>S:TEST #1 or @S:TEST #1
The various diskette housekeeping commands that you can use are explained in
Chapter 4.
2.7 Reading the Error Status
The Disk Operating System (DOS) in the 1541 monitors all disk operations. Once the
operation has been completed satisfactorily or aborted because of some error, the
DOS prepares an error status report. This report consists of an error code number, an
English language message, and the track and sector numbers of the sector where the
error, if any, occurred. Section 5.10 provides more information about the error status
messages. Appendix C contains a complete listing of the error messages and their
meanings. .
If you have DOS 5.1 active, it is very easy to read the disk drive's error status report.
To display the report all you do is type either > or @ and press the RETURN key. Try
it now. Hopefully, you should see the following display on your screen:
>
00 OK 00 00
This indicates that no errors have been encountered. Let's generate an error and try
again. Remove the diskette from your drive and enter the command >I.
When the disk drive receives the I command, it will try to initialize the diskette in the
drive (see Section 4.3). Since the drive is empty, this will produce a disk error. As soon
as you press the RETURN key the drive-active light should come on and the drive
motor will start. After a bit of whirring and clunking, the drive motor should shut off.
The drive-active LED should be flashing to signal an error. Now use the > or @ to
read the error status report. You should get the following message:
42
21 READ ERROR 18 00
This indicates a type 21 read error (no sync mark) occurred while trying to read
track 18, sector 0. Appendix C summarizes the various error status messages and
their meaning.
2.8 DOS 5.1 Commands fn Programs
Although it is not widely known, it is possible to use the DOS 5.1 commands within a
program. The only restriction is that any file names or disk commands must be
enclosed in quotation marks. Here are a few test programs for you to try:
10 PRINT "CCLR>FANCY DIRECTORY DISPLAYED
OWNJ "
20 >"*"
30 PRINT "ISN'T THAT NICE"
10 PRINT "PRESS RETURN TO INITIALIZE DIS
KETTE"
20 GET A*: IF A*OCHR*<13) GOTO 20
30 >"I"
40 PRINT "DISK HAS BEEN INITIALIZED"
10 PRINT "PRESS RETURN TO READ ERROR STA
TUS"
20 GET A*: IF A*OCHR*(13) GOTO 20
30 PRINT "ERROR STATUS ";:>
The last example illustrates two important points. A DOS 5.1 command does not have
to be on a separate program line nor does it have to be at the start of a program line.
You can put a command anywhere in your program.
The only real limitation on your use of DOS 5.1 commands in a program is that each
command must appear in the program listing exactly as you want it executed. You
cannot store file names or disk commands in string variables. The example below
illustrates what is permissible and what is not
Permissible Nonpermissible
50 >"NO:NEW DATA,D1" 45 C*="NO:NEW DATA,D1"
50 >C*
If you decide to use DOS 5.1 commands in a program, remember that you must
always have DOS 5.1 active when you use the program. Otherwise, you will get a
?SYNTAX ERROR as soon as the computer gets to the first DOS 5.1 command.
2.9 Tips on Using DOS 5.1
With DOS 5.1 you can even make use of directory listings to save some typing. Put
your 1541TEST/DEMO diskette in your drive and follow these steps:
43
Step 1: Display the directory using the >$ command like this:
>*
0 "
1541 TEST/ DEMO " ZX
2A
13
"HOW TO USE"
PRG
5
"HOW PART TWO"
PR6
4
"VIC-20 WEDGE"
PRG
1
"C-64 WEDGE"
PRG
4
"DOS 5.1"
PRG
11
"COPY/ALL"
PRG
9
"PRINTER TEST"
PRG
4
"DISK ADDR CHANGE"
PRG
4
"DIR"
PRG
6
"VIEW BAM"
PRG
4
"CHECK DISK"
PRG
14
"DISPLAY T&S"
PRG
9
"PERFORMANCE TEST"
PRG
5
"SEQUENTIAL FILE"
PRG
13
"RANDOM FILE"
PRG
558 BLOCKS FREE.
Step 2: Move the cursor up so that it is beside the name of the program that you want
to load and type either the load or load-and-run command. If you wanted to load
COPY/ALL, your display would look like this:
>*
0 "1541 TEST/DEMO " ZX 2A
13 "HOW TO USE" PRG
5 "HOW PART TWO" PRG
4 "VIC-20 WEDGE" PRG
1 "C-64 WEDGE" PRG
4 "DOS 5.1" PRG
/_ "COPY /ALL" PRG Cursor on this line.
9 "PRINTER TEST" PRG
4 "DISK ADDR CHANGE" PRG
4 "DIR" PRG
6 "VIEW BAM" PRG
4 "CHECK DISK" PRG
14 "DISPLAY T&S" PRG
9 "PERFORMANCE TEST" PRG
5 "SEQUENTIAL FILE" PRG
13 "RANDOM FILE" PRG
558 BLOCKS FREE.
Step 3: Don't worry about any extra digits in the blocks column or the file type, PRG,
behind the name. Just press the RETURN key and the load (or load and run)
command will be carried out The display will be overprinted by the "SEARCHING
44
FOR," "LOADING," and "READY." messages but this doesn't matter. If you are
using the load-only option, clear the screen to eliminate the garbage once the load
is complete.
>*
0 "
1541TEST/DEMO
ZX 2A
13
"HOW TO USE"
PRG
5
"HOW PART TWO"
PRG
4
"VIC-20 WEDGE"
PRG
1
"C-64 WEDGE"
PRG
4
"DOS 5.1"
PRG
/l
"COPY/ ALL"
PRG
9
"PRINTER TEST"
PRG
SEARCHING FOR COPY ALL" PRG
LOADINGIR" PRG
READY. VIEW BAM" PRG
"CHECK DISK" PRG
14 "DISPLAY T&S" PRG
9 "PERFORMANCE TEST" PRG
5 "SEQUENTIAL FILE" PRG
13 "RANDOM FILE" PRG
558 BLOCKS FREE.
REMINDER:
If you use this technique to load, rather than load and run a program, be sure to clear
the screen once the program has loaded. If you don't, and happen to press the
RETURN key on one of the following lines of the listing, the computer will interpret
this as a new BASIC line and add it to the program. Since the following line is not a
valid BASIC statement, your program will abort with a 7SYNTAX ERROR:
14 "DISPLAY T&S" PRG
You can use the directory listing to save a bit of typing when sending disk commands
to scratch, copy or rename a file. However, you must delete the quotation marks
around the file name and the file type (e.g., PRG) behind the name. As a result, it is
hardly worth the effort.
Once you become familiar with the commands you will find DOS 5.1 an invaluable
utility. Many people find it so helpful that they make a habit of storing a copy on each
new diskette that they format. Many thanks, Bob Fairbairn.
45
Chapter
3
THE COMMAND CHANNEL
The command channel is used to send commands to the disk drive and read the disk
drive's error status report. You can use DOS 5.1 to carry out these tasks from immedi-
ate mode. However, if you will be writing programs that use the disk drive, you will
need to know how to use the command channel.
Even if you are a more experienced 1541 user you should skim through this chapter.
There are a number of interesting tidbits, including four different ways to read the
error status from immediate mode without using DOS 5.1!
3.1 The Command Channel
The command channel is a special communications channel between your computer
and disk drive. It is used to:
1. Send diskette housekeeping commands to the disk drive.
2. Send direct-access commands to the disk drive.
3. Input the error status from the disk drive.
This chapter covers how to send diskette housekeeping commands and how to input
the error status. The direct-access commands are not required for normal file handl-
ing. They are used primarily in specialized disk utility programs. As a result, they
are not discussed in this book. However, many of the disk utility programs listed in
this book make use of direct-access commands. The more advanced book, Inside
Commodore DOS, by Richard Immers and Gerald G. Neufeld, (DATAMOST, 1984),
discusses how to use these commands effectively.
3.2 Opening the Command Channel
To establish communications between your computer and the disk drive you use an
OPEN statement. Since the command channel is channel number 15, the OPEN
statement will look like this:
SYNTAX:
OPEN file*, device*, cmd chnl#
EXAMPLES:
OPEN 15,8,15
OPEN 15,9,15
47
WHERE:
file# = the logical file number (any integer 1-127). Often 15 is used
as a file# for the command channel because it is easy to
remember that this is the command channel. In this book
the command channel will always be file# 15.
device# = the device number. Normally a 1541 is device number
eight. A 1541's device# can be changed either temporarily
or permanently (see Sections 5.7 and 12.3).
cmd chnl# = 15, the command channel number. The command channel
is always channel* 15. Channels 0 and 1 are reserved for
LOAD and SAVE. Channels 2-14 are for file handling.
You can issue an OPEN statement either by typing it in and pressing the RETURN
key (immediate mode) or from within a BASIC program. When you open the
command channel, nothing will appear to happen.
3.3 Sending Disk Commands
A disk command is a message to the 1541's Disk Operating System (DOS) that tells it
to do a particular task such as formatting a diskette or renaming a file. Disk com-
mands are sent to the disk drive using a PRINT# statement. PRINT# is a BASIC
statement and is similar to the more familiar PRINT statement. It looks like this:
SYNTAX:
PR I NT# * i 1 e# , "command "
examples:
print* 15, "no: test disk,t3"
PRINT#15, "10"
PRINT#15,"V0"
WHERE:
file# = the logical file number used when you opened the command
channel.
command = the diskette housekeeping command to be sent to the disk
drive. The commands to use are described in detail in
Chapter 4.
NOTES:
1. PRINT# is a single BASIC keyword. You may not put a space between PRINT
and the # sign. Spaces following the # sign are optional.
2. Use the correct abbreviation. You may abbreviate PRINT# by typing P followed
by a shifted R (pR in upper/lower case mode). Do not use ?# as an abbreviation.
This appears to be correct in a program listing but causes a ?SYNTAX ERROR.
48
You can issue a PRINT# statement either by typing it in and pressing the RETURN
key (immediate mode) or by including it in a BASIC program. When you send a
command, the disk drive will interpret the command and carry it out. Most commands
cause the drive-active LED to turn on and the drive motor to start. It may take
several minutes for the command to be carried out. While the disk drive is carrying
out the command, the computer is free. You can play a game, type in a new program
or edit an existing program. However, if you try to access the disk drive, the computer
will appear to hang up until the disk drive has completed the previous command.
Once the command has been completed, the drive motor will stop and the drive-active
LED should go out.
If the drive-active LED begins to flash ON and OFF regularly, an error has occurred.
Either the command was incorrect or a disk error has occurred. Read the disk drive's
error status to determine the cause of the error (see Sections 3.6 and 3.7).
3.4 Combining OPEN and PRINT# Statements
An alternate form of the OPEN command allows you to send a single disk command
to the disk drive when you open the command channel. This combined form looks
like this:
SYNTAX :
OPEN -file#, device*, cmd chnl#„ "disk command"
EXAMPLES:
OPEN 15,8, 15, "NO: GOOD GAMES, Gl"
OPEN 15,8, 15, "SO: TEST FILE"
OPEN 15,8, 15, "VO"
OPEN 15,8, 15, "10"
WHERE:
file# = the logical file number used when you opened the command
channel.
device# = the device number. Normally a 1541 is device number
eight. A 1541's device# can be changed either temporarily
or permanently (see Sections 5.7 and 12.3).
cmd chnl# = 15, the command channel number. The command channel
is always channel# 15. Channels 0 and 1 are reserved for
LOAD and SAVE. Channels 2-14 are for file handling.
command = the disk housekeeping command to be sent to the disk drive.
The commands to use are described in detail in Chapter 4.
This form of the OPEN statement is handy when you have just a single disk command
to send. If you get a ?FILE OPEN message when you use this form, this file# is
already open. When this happens, it is usually easiest to close the file by typing
CLOSE 15 and pressing the RETURN key (see Section 3.8). Now, move the cursor up
to the OPEN statement, and press the RETURN key to reopen the command channel
and send the command.
49
3.5 Reading the Command Channel
The most common reason for reading the command channel is to access the 1541's
error status message. However, you can also use it to read the contents of the disk
drive's RAM or ROM memory by using a direct access MEMORY-READ (M-R)
command. Since this is not required for most operations, we will focus on reading the
error status.
The Disk Operating System (DOS) in the 1541 monitors all disk operations and
generates an error status report when the operation is completed or aborted. The
report consists of four parts:
1. An error code number.
2. An English language error message.
3. The track number where the error, if any, occurred.
4. The sector number where the error, if any, occurred.
The various possible messages and their meanings are explained in detail in
Appendix C.
You access the disk drive's error status report by reading the command channel.
Normally an' INPUTS statement is used to read the report. However, you can also use
a GET# statement. Unfortunately, both INPUT# and GET# can only be used within
a BASIC program. They may not be used in immediate mode. However, you can use
simulated GET# and INPUTS commands to read the error status from immediate
mode as indicated in Section 3.7.
In order to read the report you need to know a bit about the INPUTS and GETS
statements. Only those characteristics of these statements that apply to reading the
error channel are mentioned here. For more details regarding these statements, see
Section 5.10. The usual format for the INPUTS statement is:
SYNTAX:
INPUT* file*, var list
EXAMPLES:
INPUT* 15,EN,EM*,ET,ES
INPUT* 5,NAM*(K)
WHERE:
fileS = the file number used when you opened the command
channel.
var list = a list of one or more variable names. If more than one
name appears, the names must be separated by commas.
NOTES:
1. INPUTS is a single BASIC keyword. You may not put a space between INPUT
and the S sign. Spaces following the S sign are optional.
50
2. Use the correct abbreviation. You may abbreviate INPUT# by typing I followed
by a shifted N (iN in upper/lower case mode).
3. INPUT# reads one or more complete numeric or string values at one time. Since
the error status report consists of one string and three numeric values, the INPUT#
statement can read and assign all these values quickly and easily. The INPUT#
statement is the preferred way to read the drive's error status.
The usual format for the GET# statement is:
SYNTAX:
SET# file#, var list
EXAMPLES:
GET# 15, A*
6ET# 5,J*,K*,L*
WHERE:
file# = the file number used when you opened the command
channel.
var list = a list of one or more variable names. If more than one
name appears, the names must be separated by commas.
NOTES:
1. GET# is a single BASIC keyword. You may not put a space between GET and the
# sign. Spaces following the # sign are optional.
2. There is no abbreviation for GET#.
3. GET# stores only one character at a time in the named variable(s). As a result,
GET# is not normally used to read the disk drive's error status.
3.6 Reading the Error Status in a Program
It is a good idea to read the error status after every disk operation in your program to
be sure that the operation was completed successfully. You may use either an INPUTS
or a GET# statement to read the status. However, since you will probably want to
interpret the status report to determine whether or not to abort the program, you will
usually want to use the INPUT# form. The GET# form is only useful for displaying
the disk status on the screen without interpreting it.
No matter which form you use, you will probably want to write the error status check
as a subroutine. This allows you to check the disk status quickly and easily.
INPUT# form of reading the disk drive's error status report:
51
10 OPEN 15,8,15
150 PRINT#15, "NO: TESTER, XY"
160 GOSUB 500
• •■••■
'■■•■■•
490 CLOSE 15
49? END
500 REM SUB TO CHECK STATUS
510 INPUT#15,EN,E*,ET,ES
520 IF EN<20 THEN RETURN
530 PRINT "FATAL DISK ERROR"
540 PRINT en;e*;et;es
550 CLOSE 15
560 END
Open command channel.
Program continues.
Disk operation.
Check disk drive status.
Program continues.
Close command channel.
Read error status.
Everything is OK.
Trouble.
Print status report.
Close all disk files.
Abort program.
GET# form of reading the disk drive's error status report:
10 OPEN 15,8,15
150 PRINT#15,"N0:TESTER,XY"
160 GOSUB 500
490 CLOSE 15
499 END
500 REM SUB TO PRINT STATUS
510 GET#15,A*
520 PRINT A*;
530 IF A*OCHR*<13> GOTO 510
540 RETURN
Open command channel.
Program continues.
Disk operation.
Print disk drive status.
Program continues.
Close command channel.
Read one character.
Print one character.
Read till carriage return.
3.7 Reading the Error Status in Immediate Mode
Normally, it is impossible to read the error status in immediate mode. The reason is
that the INPUT# and GET# statements only work when they are included in a
program. If you try to use them in immediate mode, you will get an 7ILLEGAL
DIRECT ERROR. However, it is possible to simulate program mode, or use the
Commodore 64's ROM routines to simulate these commands and read the error
report from direct mode as indicated below.
Simulated GET# for reading the error status in immediate mode:
To read the error status while in immediate mode, type in the following commands.
Remember to press RETURN after each line.
CLOSE 15
P0KE58, 0: 0PEN15, 8, 15: F0RK=1T030: GET#15, A
*: ?a*; : ifst=othennext
52
or
CL0SE15: 0PEN15, 8, 15
F0RK=1T030: SYS (43909) 15, A*:? A* 5 : IFST=OTH
ENNEXT
In the first version, the POKE command causes the computer to think it is running a
program. This allows the use of the normal GET# statement. If you are sure that file#
15 is not open, you can eliminate the initial CLOSE 15 line. However, if file# 15
is already open, you must close it (see Section 3.8) and then reopen it as part of the
GET# line.
In the second version, the SYS command makes use of the Commodore 64's GET#
ROM routine but enters it after the check for immediate mode. In this version you
can eliminate the first line if file# 15 is already open to the command channel.
Simulated INPUT# for reading the error status in immediate mode:
An alternate way to read the error status while in immediate mode is to use a
simulated INPUT#. As before you can use either of two versions. Take your pick.
Remember to press RETURN.
CLOSE 15
P0KE58,0: POKE 198, l:0PEN15,8, 15: INPUT#15,
e,e*,t,s:?e;e*;T;s
Note the space between the comma following INPUT# 15 and the E. This space must
be there.
or
CLOSE 15
P0KE58, O: 0PEN15, 8, 15: SYS (43941 > 15, E, E«,
t,s:?e;e*;t;s
Note the space between the comma following SYS(43941)15 and the E. This space
must be there.
If you are sure file# 15 is not open, you may omit the CLOSE 15 line with either
version. Since the check for program mode is deep within the INPUT# ROM routine,
both versions use the POKE to simulate program mode and must include the OPEN
command. Since the first version uses only standard BASIC commands, you will
probably find it easier to remember.
3.8 Closing the Command Channel
Once you have finished using the command channel you should close it. You use the
BASIC statement CLOSE to do this. It looks like this:
53
syntax:
CLOSE file#
EXAMPLES:
CLOSE 15
CLOSE 15
WHERE:
file# = the file number used when you opened the command
channel.
NOTES:
1. Closing the command channel closes all files in the 1541. When you close the
command channel, all disk files that are open will be closed automatically in the
disk drive. Because of this, you should OPEN the command near the beginning of
any program you write and leave it open until the end of your program. See the
HINT below for a useful application of this. CAUTION: Although the files will be
closed properly in the disk drive, they will still be open in the computer.
2. Loading, saving, running, or editing closes files in the computer. Whenever you
LOAD, SAVE or RUN a program, all files that are open in the computer,
including the command channel, will be closed. Adding to or editing a program
will close all the files in the computer except the command channel. CAUTION:
Although the files will be closed in the computer, they will still be open in the
drive. If this happens, you may end up with one or more unclosed disk files. Use
the HINT below to prevent this.
You can issue a CLOSE statement either by typing it in and pressing the RETURN
key (immediate mode) or by including it in a BASIC program. When you CLOSE the
command channel, you will not see any obvious effect unless some other disk files are
also open. If one or more other files are open (the drive active light will be on), the
drive motor will turn on briefly as the files are closed. After a few seconds, the drive
motor will stop and the drive-active LED should go out. If the drive-active LED
begins flashing, an error has occurred. Read the drive's error status report (see
Sections 3.6 and 3.7) to find the cause of the error.
HINT:
Preventing improperly closed disk files. Have you ever had a program that aborted
with a 7SYNTAX ERROR and left a disk file open? As long as you spot the red
drive-active light glowing before you edit that offending line, everything is fine; the
file is still open in both the computer and the disk drive. All you do is issue the
appropriate CLOSE statement from immediate mode and the file will be closed
properly. If you forget the file#, you can either LIST the program and find it or use
the command channel fix below.
However, most of us are in such a rush to fix that offending line that we don't notice
that the drive-active light is on until after we have edited the line. Once you do any
editing, you have a problem. The computer will have closed the file automatically as
54
soon as you did any editing, but the file is still open in the drive. Issuing a CLOSE
command won't work because the file is already closed in the computer. The solution
is to enter this line and press RETURN.
CL0SE15:0PEN15,8, 15:CL0SE15
This works because of the special characteristics of the command channel. You can
OPEN it without affecting any other active disk files. However, when you CLOSE it,
any other active disk files are closed. The first CLOSE 15 will close all the disk files if
file# 15 happens to be open to the command channel. The rest of the command opens
the command channel and then closes it immediately. The final CLOSE ensures that
all active files will be properly closed.
55
Chapter
4
DISKETTE HOUSEKEEPING
NOTE:
This chapter is designed for easy reference. As a result, each section must be indepen-
dently readable and understandable. While this format may appear awkward and
repetitious in a complete reading, users will rapidly discover that it saves much page
flipping.
Unless you plan on only using commercially recorded software, you will have to do
some housekeeping to maintain your collection of diskettes. As you saw in Chapter 1,
you even have to do some housekeeping before you can save anything on a new, blank
diskette; it must be formatted (newed) first. This chapter is designed to take some of
the mystery out of diskette housekeeping. It not only tells you what to do, it also tells
you what happens to your diskette when you issue a housekeeping command.
Diskette housekeeping involves sending commands to the disk drive over the command
channel. You can either use the command channel directly (see Chapter 3) or use
DOS 5.1 to make your job easier (see Chapter 2). In immediate mode you can use
either method. If you will be building diskette housekeeping into a program, you will
have to use the command channel directly.
There are a variety of tasks that are considered diskette housekeeping. They include
the following:
1. Formatting a new diskette (full new).
2. Reusing a diskette (short new).
3. Initializing a diskette.
4. Scratching a file.
5. Renaming a file.
6. Copying a file.
7. Combining two or more files.
8. Validating a diskette.
This chapter explains how to do each of these tasks and describes what actually
happens to your diskette as the command is carried out. If you feel somewhat lost
after reading the chapter, don't panic! The chapter concludes with a listing of a
menu-driven BASIC program that makes diskette housekeeping a breeze.
A few additional, non-standard diskette housekeeping tasks such as creating peculiar
directory entries and changing a diskette's name are discussed in Section 13.8.
57
4.1 Formatting a New Diskette
When you purchase a new diskette there is nothing recorded on it. Before you can use
it for storing programs or data files, the diskette must be formatted (newed). Despite
what you may have heard from your Apple owning friends, this is nottitie same as
initializing a diskette. In Commodore jargon, initializing a diskette means reading
some directory information from the diskette and storing it in the disk drive's memory
(see Section 4.3). Formatting or newing a diskette involves recording a pattern of
information on each of 35 concentric tracks. This information allows the drive to tell
which track, as well as which part of the track, it is on. The formatting process takes
about 80 seconds to complete. Once formatting is completed, any information pre-
viously recorded on that diskette is permanently lost
A NEW command is used to tell the disk drive to format a diskette. This is not the
same as the NEW command in BASIC. This NEW command is a disk command and
looks like this:
General -form:
NEWO:disk name, id
Alternate form:
IM:disk name, id
Examples:
ne wo: good GAMES #1,G1
n:word PROCESSING, Wl
WHERE:
NEW = is the disk command. The command NEW may be abbre-
viated as N. The zero is the drive number. It is a holdover
from Commodore's dual drives. Its use is optional on the
1541 disk drive.
disk name = is the name that will appear on the first line of the direc-
tory listing. The name may be up to 16 characters long.
Any letters, numbers, or graphics characters, except a
comma (,), a question mark (?) or an asterisk (*) may be
used. Using reverse-field characters in names is not
recommended.
id = is a two character disk ID. This appears following the disk
name at the top of the directory listing. Any letters,
numbers, or graphics characters may be used.
CAUTION:
Use a different ID for each diskette you format! It is a good idea to develop a
systematic numbering system for your diskettes as soon as you buy your first box of
diskettes. This will save a lot of renaming, renumbering and recopying later. For
example, you might number your word processing diskettes Wl, W2,..., your games
diskettes Gl, G2,..., etc..
58
Sending a NEW command using DOS 5.1:
SYNTAX:
>NO:disk name, id
ALTERNATE:
©NO: disk name, id
EXAMPLES:
>NO:GOOO GAMES #1,61
@N:WORD PROCESSING, Wl
Sending a NEW command using the command channel:
SYNTAX:
OPEN 15,8,15
PRINT#15,"NO:disk name, id"
CLOSE 15
ALTERNATE:
OPEN 15,8, 15, "NO: disk name, id": CLOSE 15
EXAMPLE #1: (immediate mode)
OPEN 15,8,15
PR I NT# 1 5 , " NO : C-64 PROGRAMS , 64 "
CLOSE 15
EXAMPLE #2: (immediate mode)
OPEN 15,8, 15, "NO: SUPER STUFF, SS": CLOSE 15
EXAMPLE #3: (program mode)
lOO OPEN 15,8,15
■ ■ ■ m m
• • ■ ■ ■
360 PRINT#15, "N:CLUB MAIL LIST, CI"
■ ■ m m m
500 CLOSE 15
EXAMPLE #4: (program mode)
100 OPEN 15,8,15
• ■ ■ ■ ■
■ • m m m
350 FI*="N0:C-64 PROGRAMS, 64"
360 PRINT#15,FI*
: : : : :
500 CLOSE 15
EXAMPLE #5: (program mode)
100 OPEN 15,8,15
" ■ • m m
250 INPUT "NEW DISK'S NAME";DN*
260 INPUT"NEW DISK'S ID";DI*
270 PRINT#15, "N: "+DN*+" , "+DI*
■ « m m m
500 CLOSE 15*
59
Examples 4 and 5 illustrate how you can use string variables to store either the entire
disk command (4) or parts of a disk command (5). You can use this technique in either
immediate or program mode as long as you use the command channel. You cannot
use this technique when using DOS 5.1.
What Happens When You Issue a NEW Command?
When you send a NEW command to the disk drive, the disk operating system in the
drive interprets (parses) the command. If the syntax is incorrect the command is
rejected, the red drive-active light begins flashing and a SYNTAX ERROR report is
generated (error number 30, 31, 32, 33, 34, or 39).
If the syntax is correct, formatting begins. The red drive-active LED turns on, the
drive motor starts up and the drive's stepper motor moves the record/play head to the
outer edge of the diskette (track 1). The head is moved outward a total of 45 tracks.
When it reaches track 1 a cam on the drive pulley hits a mechanical stop. It is the cam
hitting the stop that causes the clattering noise so familiar to 1541 owners. For rather
obvious reasons this is known as doing a bump to track 1.
The record/play head is now stationary and the diskette is rotating at 300 rpm. As the
diskette makes one complete rotation a circular strip of the diskette passes over the
record/play head. This circular strip, about l/64th of an inch wide, is called a track.
First, any existing information recorded on this strip (track) is erased. Then a new
pattern of information is recorded. This new information subdivides the track into a
number of sectors. The diagram below shows how the tracks and sectors are arranged
and numbered on a 1541 formatted diskette.
The number of sectors on a track is not constant. The tracks near the outer edge of the
disk are longer and are divided into more sectors. The chart below shows how the
number of sectors varies according to the track:
60
Zone
Track Numbers
Sector Numbers
Total Sectors
1
ltol7
0to20
21
2
18 to 24
0tol8
19
3
25 to 30
0tol7
18
4
31 to 35
0tol6
17
Each sector on a track consists of two parts, an identifying header and a data block
containing 256 bytes (characters) of data. Here is an overview of what a short segment
of a track looks like:
SYNC
MARK
Header of
Sector #0
Header
Gap
SYNC
MARK
Data block
of Sector #0
Tail
Gap
SYNC
MARK
Header of
Sector ttl
Header
Gap
SYNC
MARK
Note that both the header and the data block begin with a special character called a
SYNC MARK and end with a gap. The SYNC MARK alerts the disk operating
system that important information is coming up. The gap allows the drive some time
to get ready for the next SYNC MARK. During formatting both the header informa-
tion and a dummy data block are recorded for each of the sectors that make up a
track. When you save a program or data file on the diskette, the original data blocks
will be overwritten by the information to be saved. However, the header information
is never altered unless you reformat the diskette.
Once all the headers and data blocks on the track have been recorded, they are read
back and checked to make sure they have been recorded properly. If everything
checks out, the head is moved inward one track (you'll hear a click as the stepper
motor moves the head), and the process is repeated. If an error is detected, the disk
operating system will try formatting that track again. You can get some idea of how
many attempts it took by noting the time between clicks. If there is some flaw on the
diskette that prevents formatting a track correctly, the disk operating system will
eventually give up and report a 21, READ ERROR.00,00. Why on track 0, sector 0?
Just because.
When all 35 tracks have been formatted, the stepper motor will move the head
outward to track 18. The disk name, disk ID, and a blank Block Availability Map
(BAM) will be recorded on track 18, sector 0. A blank directory will be recorded on
track 18, sector 1. Once this has been done, formatting is complete. The drive-active
LED and the drive motor are switched off.
4.2 Reusing a Diskette
Eventually you will want to erase everything off one of your diskettes and reuse it for
something else. One way to do this is to format the entire diskette again as we did in
Section 4.1 for a blank diskette. A quicker method is to simply erase the existing disk
name, directory, and Block Availability Map (BAM) and leave the rest of the diskette
alone. As this takes only a few seconds to do and you use a variation of the NEW
command to do it, this is often referred to as a short NEW. Note that the command is
very similar to an ordinary NEW, the only difference is that you omit the disk ID.
61
General -form:
NEWO:disk name
Alternate -form:
N:disk name
Examples:
no: good games #1
n:word processing
WHERE:
NEW = is the disk command. The command NEW may be abbre-
viated as N. The zero is the drive number. It is a holdover
from Commodore's dual drives. Its use is optional on the
1541 disk drive.
disk name = is the name that will appear on the first line of the direc-
tory listing. The name may be up to 16 characters long.
Any letters, numbers or graphics characters, except a
comma (,), a question mark (?) or an asterisk (*) may be
used. Using reverse-field characters in names is not
recommended.
Sending a short NEW command using DOS 5.1:
SYNTAX:
>N:disk name
alternate:
©NO: disk name
examples:
>no:good games #1
@n:word processing
Sending a short NEW command using the command channel:
SYNTAX:
OPEN 15,8,15
PRINT#15,"N:disk name"
CLOSE 15
ALTERNATE:
OPEN 15, 8, 15, "NO: disk name" : CLOSE 15
EXAMPLE #l: (immediate mode)
OPEN 15,8,15
PR I NT# 1 5 , " N : C-64 PROGRAMS "
CLOSE 15
62
EXAMPLE #2: (immediate mode)
OPEN 15,8, 15, "NO: SUPER STUFF": CLOSE 15
EXAMPLE #3: (program mode)
100 OPEN 15,8,15
■ ■ • • ■
• ■ • ■ •
360 PRINT#15, "N:CLUB MAIL LIST"
: z : : ■
500 CLOSE 15*
EXAMPLE #4: (program mode)
100 OPEN 15,8,15
: : : : :
350 FI*="N:C-64 PROGRAMS"
360 PRINT#15,FI*
■ ■ « • »
500 CLOSE 15*
EXAMPLE #5: (program mode)
100 OPEN 15,8,15
• ■ • ■ ■
■ • ■ ■ ■
250 INPUT "NEW DISK'S NAME";DN*
270 PRINT#15, "N: "+DN*
■ • • • •
■ • • • ■
500 CLOSE 15
Examples 4 and 5 illustrate how you can use string variables to store either the entire
disk command (4) or a part of the disk command (5). You can use this technique in
either immediate or program mode as long as you use the command channel. You
cannot use this technique when using DOS 5.1.
What happens when you issue a short NEW command?
When you send a NEW command to the disk drive, the disk operating system in the
drive interprets (parses) the command. If the syntax is incorrect, the command is
rejected, the red drive-active light begins flashing and a SYNTAX ERROR report is
generated (error number 30, 31, 32, 33, 34, or 39).
If the syntax is correct, the command is carried out. As always, the red drive-active
LED turns on and the drive motor starts. This time, however, the head moves directly
to track 18. The new disk name, the old disk ID (it hasn't been changed), and a blank
Block Availabilty Map (BAM) will be recorded on track 18, sector 0. A blank
directory will be recorded on track 18, sector 1. The remainder of the diskette is
unaffected. Once these two sectors have been modified, the drive-active LED and the
drive motor are switched off. Note that this is much different than what happens
when a full reformatting takes place. In that case, all the sectors on the diskette were
rewritten and any information previously stored on the diskette was permanently
lost.
63
WARNING:
If you have been encountering disk errors when using a particular diskette, do not use
a short NEW on it. If you do, the errors will remain. The only hope for a diskette that
contains read errors is to copy as much information as possible onto another diskette
and then reformat the entire error ridden diskette using a full NEW as described in
Section 4.1.
NOTE:
Since only two sectors on the diskette are modified during a short NEW, it is possible
to recover the information stored on the diskette if you issued a short NEW command
by mistake. Section 11.4 explains how to do this.
4.3 Initializing a Diskette
In Commodore jargon, initializing a diskette means reading a copy of a diskette's
Block Availability Map (BAM) from track 18, sector 0 into the disk drive's RAM
memory. Normally the drive does this automatically whenever you insert a diskette
and perform any disk operation. However, it is a good idea to initialize the diskette
whenever you switch diskettes and whenever a disk error occurs. This ensures that
the drive will correctly recognize the diskette. An initalize command looks like this:
General form:
INITIALIZEO
Alternate -form:
I
Examples:
10
I
WHERE:
INITIALIZE = the disk command. The command INITIALIZE may
be abbreviated as I. The zero is the drive number. It
is a holdover from Commodore's dual drives. Its use
is optional on the 1541 disk drive.
Sending an INITIALIZE command using DOS 5.1:
SYNTAX:
>I0
ALTERNATE:
@I0
examples:
>I0
ei
64
Sending an INITIALIZE command using the command channel:
SYNTAX:
OPEN 15,8,15
PRINT#15, "10"
CLOSE 15
ALTERNATE:
OPEN 15,8,15, " 10": CLOSE 15
EXAMPLE #l: (immediate mode)
OPEN 15,8,15
PRINT#15, "10"
CLOSE 15
EXAMPLE #2: (immediate mode)
OPEN 15,8, 15, "10": CLOSE 15
EXAMPLE #3: (program mode)
100 OPEN 15,8,15
■ • • ■ •
• • • • •
360 PRINT#15, "I"
■ ■ • • m
500 CLOSE 15
EXAMPLE #4: (program mode)
100 OPEN 15,8,15
» • ■ • •
■ ■ • • ■
350 FI*="IO"
360 PRINT#15,FI*
■ « ■ • *
• • • • •
500 CLOSE 15
Example 4 illustrates how you can use string variables to store the disk command.
You can use this technique in either immediate or program mode as long as you use
the command channel. You cannot use this technique when using DOS 5.1.
What happens when you issue an INITIALIZE command?
When you send an INITIALIZE command to the disk drive, the disk operating
system in the drive interprets (parses) the command. If the syntax is incorrect, the
command is rejected, the red drive-active light begins flashing and a SYNTAX
ERROR report is generated (error number 30, 31, 32, 33, 34, or 39).
If the syntax is correct, initialization begins. As always, the red drive-active LED
turns on and the drive motor starts. The head moves directly to track 18. The data
block of track 18, sector 0 is read into the disk drive's RAM memory (from $0700-
07FF). The disk drive's copy of the diskette's ID (in memory location $12/$13) is also
updated. Once this happens the drive-active LED and the drive motor are turned off.
65
NOTES:
Occasionally you may find that you get a 29, DISK ID MISMATCH,00,00 error when
you initialize a diskette. Don't panic, just initialize the diskette again and the error
should go away. If you get a 21, READ ERROR,18,00, you are probably trying to
initialize a brand new, unformatted diskette. If you are sure the diskette is formatted
and still get read errors, you have serious problems. See Section 11.5 for possible
solutions.
4.4 Scratching a File
When you no longer need a program or data file, you can scratch it This will eliminate
the file's name from the directory and free the sectors (blocks) that were used to store
the program or data file. This allows you to save new information there. A SCRATCH
command looks like this:
General -form:
SCRATCHO:file name
Alternate -form:
S:-File name
Examples:
SO: OLD JUNK
s:test ??.BAS
WHERE:
SCRATCH = is the disk command. The command SCRATCH may be
abbreviated as S. The zero is the drive number. It is a
holdover from Commodore's dual drives. Its use is option-
al on the 1541 disk drive.
file name = is the name of the file to be scratched. Wild cards (? and*)
may be used in the name.
WARNINGS:
1. More than one file may be scratched. If you use wild cards in specifying the name
of the file to be scratched, the resulting pattern may match more than one file
name in the directory. If this happens, all the tiles whose names match the pattern
will be scratched.
2. Do not scratch an unclosed file. You should not scratch a file that has an * beside
its file type in the directory (e.g., *PRG, *SEQ, or *USR). The * indicates that this
is a file that has not been closed properly. These files should be eliminated by
using a VALIDATE command (see Section 4.8). If you accidentally SCRATCH an
unclosed file, VALIDATE the diskette as soon as possible.
NOTE:
When you scratch a file, you create a vacancy in the directory file. The next time you
save a program or store a data file on this diskette the new name will not be added to
the end of the directory. It will fill the vacancy in the directory.
Sending a SCRATCH command using DOS 5.1:
SYNTAX:
>S:file name
ALTERNATE:
©SO: file name
EXAMPLES:
>SO: TAXES 1981
@S:WP/»
Sending a SCRATCH command using the command channel:
SYNTAX :
OPEN 15,8,15
PRINT#15,"S0:file name"
CLOSE 15
ALTERNATE:
OPEN 15,8, 15, "SO: file name": CLOSE 15
EXAMPLE #1: (immediate mode)
OPEN 15,8,15
PRINT#15, "SO: GAME VERSION 1"
CLOSE 15
EXAMPLE #2: (immediate mode)
OPEN 15,8, 15, "SO: BUNCHA*": CLOSE 15
EXAMPLE #3: (program mode)
lOO OPEN 15,8,15
360 PRINT#15, "SO
• a a a
• • ■ a
500 CLOSE 15
V7/SAME.P"
EXAMPLE #4: (program mode)
100 OPEN 15,8,15
a a a a a
■ • a a a
350 FI*="SO:UNPAID*"
360 PRINT#15,FI*
a • a a a
a a a a a
500 CLOSE 15
EXAMPLE #5: (program mode)
100 OPEN 15,8,15
a a a a a
a a a a a
250 INPUT "NAME OF FILE TO SCRATCH" ; DN*
260 PRINT#15, "S: "+DN*
a a a • a
a a a a a
500 CLOSE 15
67
Examples 4 and 5 illustrate how you can use string variables to store either the entire
disk command (4) or portions of the command (5). You can use this technique in either
immediate or program mode as long as you use the command channel. You cannot
use this technique when using DOS 5.1.
What happens when you issue a SCRATCH command?
When you send a SCRATCH command to the disk drive, the disk operating system in
the drive interprets (parses) the command. If the syntax is incorrect, the command is
rejected, the red drive-active light begins flashing and a SYNTAX ERROR report is
generated (error number 30, 31, 32, 33, 34, or 39).
If the syntax is correct, the scratching process begins. As usual, the red drive-active
LED turns on and the drive motor starts. The head moves directly to track 18 and
begins reading the directory file (18-1, 18-4, and so on). When it finds a name that
matches the one specified in the SCRATCH command, it checks the file type byte for
this entry. What happens next depends on the value of the file type byte. If this is a
normal SEQ, PRG, USR or REL file, the file type byte is set to $00 (this entry will no
longer appear in a directory) and the DOS begins tracing the file across the surface of
the diskette. The diskette's Block Availability Map (BAM) is modified to indicate that
the sectors the file occupied are now free and may be reused. Once the end of the file is
reached, the head returns to track 18 and continues reading the directory looking for
names that match the file name specified. Once the end of the directory is reached,
the revised BAM is written to track 18, sector 0. The operation is complete. The
drive-active LED and the motor are switched off.
NOTE:
Since scratching a file does not erase the information from the diskette, it is possible
to unscratch a file provided that you have not recorded any new information on the
diskette after you scratched the file. If you scratch a file by accident, remove the
diskette from the drive and read through Section 11.2.
Now that you know what normally happens, let's look at the more unusual situations.
If the file is locked (bit six of the file type byte is a one), the scratch operation
terminates. You can't scratch a locked file. If the file is unclosed (bit four of the file
type byte is a zero), all that happens is that the file type byte is set to $00. An unclosed
file is not traced across the diskette on a 1541 drive! As a result, the blocks that held
the file will not be freed in the BAM. You must VALIDATE the diskette to free these
blocks!
4.5 Renaming a File
You may change the name of any file that appears in a diskette's directory listing
using the RENAME command. Only the name that appears in the directory is
changed, the actual program or data file is unaffected. A RENAME command looks
like this:
68
General -form:
RENAMEOmew name=old name
Alternate form:
Rznew name-old name
Ex amp les:
RO: JUNK=TREASURES
R: YESTERDAY=TODAY
WHERE:
RENAME = the disk command. The command RENAME may be
abbreviated as R. The zero is the drive number. It is a
holdover from Commodore's dual drives. Its use is op-
tional on the 1541 disk drive.
new name = the revised name that you want to appear in the directory
listing. The name may be up to 16 characters long. Any
character except asterisks, colons, quotation marks,
question marks or commas, may be used in the name.
The use of reverse-field characters in names is not
recommended.
old name = the name that currently appears in the directory listing.
The old name must be spelled exactly as it appears. Wild
cards (? and *) may not be used.
Sending a RENAME command using DOS 5.1:
SYNTAX:
>RO:new name=old name
ALTERNATE:
@R:new name=old name
EXAMPLES:
>ro:past DUE=CURRENT BILLS
@r:past affairs=love letters
Sending a RENAME command using the command channel:
SYNTAX:
OPEN 15,8,15
PRINT#15, "R:new name=old name"
CLOSE 15
ALTERNATE:
OPEN 15,8, 15, "RO: new name=old name": CLOSE 15
69
EXAMPLE #1: (immediate mode)
OPEN 15,8,15
PRINT#15,"R:90 DAYS LATE=60 DAYS LATE"
CLOSE 15
EXAMPLE #2: (immediate mode)
OPEN 15,8, 15, "RO:WIFE=GIRL FRIENDM:CLOSE 15
EXAMPLE #3: (program mode)
100 OPEN 15,8,15
360 PR INT#15,"R: FINAL GAME=VERSION 27"
■ m ■ ■ •
■ m m • •
500 CLOSE 15
EXAMPLE #4: (program mode)
lOO OPEN 15,8,15
■ • ■ ■ ■
350 FI*="RO:PAID BILLS=UNPAID BILLS"
360 PRINT#15,FI*
• • • ■ »
■ • • • •
500 CLOSE 15
EXAMPLE #5: (program mode)
100 OPEN 15,8,15
• • • • •
250 INPUT"NAME OF FILE TO RENAME"; ON*
2AO INPUT "NEW FILE NAME";NN*
270 PR I NT# 1 5 , " R : " +NN*+ " ■ " +0N*
• • « ■ •
• ■ • • ■
500 CLOSE 15
Examples 4 and 5 illustrate how you can use string variables to store either the entire
disk command (4) or portions of the command (5). You can use this technique in either
immediate or program mode as long as you use the command channel. You cannot
use this technique when using DOS 5.1.
What happens when you issue a RENAME command?
When you send a RENAME command to the disk drive, the disk operating system in
the drive interprets (parses) the command. If the syntax is incorrect, the command is
rejected, the red drive-active light begins flashing and a SYNTAX ERROR report is
generated (error number 30, 31, 32, 33, 34, or 39).
If the syntax is correct, renaming begins. As usual, the red drive-active LED turns on
and the drive motor starts. The head moves directly to track 18 and begins reading
the directory file (18-1, 18-4, and so on). It reads through the entire directory checking
for both the old name and the new name. If the old name is not found, the command is
aborted and the drive reports a 62, FILE NOT FOUND,00,00 error. If the new name
matches an existing name in the directory, the command is aborted and the drive
70
reports a 63, FILE EXISTS.00,00 error. If everything looks OK, the old name in the
directory is replaced by the new name. Once the modified directory sector is recorded
on disk, the drive-active LED and the motor are switched off.
4.6 Copying a File
You may make a backup copy of any file, including a relative file, that appears in a
diskette's directory listing using the COPY command. Unfortunately, this command
is of little practical use because the backup copy must always be on the same diskette
as the original file. If you want to make a copy onto another diskette, you must use the
COPY/ALL program on the 1541TEST/DEMO diskette or the FAST COPY program
at the end of Chapter 10. A COPY command looks like this:
General •form:
COPYO: backup name-O: original name
Alternate form:
C: backup name=original name
Examples:
CO: OLD JUNK/2=0LD JUNK
C: BUNCH ABACKUP=BUNCHASTUFF
WHERE:
COPY = the disk command. The command COPY may be
abbreviated as C. The zero is the drive number. It is a
holdover from Commodore's dual drives. Its use is
optional on the 1541 disk drive.
backup name = the name of the new backup copy as it should appear
in the directory. The name may be up to 16 characters
long. Any characters except asterisks, colons, quota-
tion marks, question marks or commas, may be used
in the name. The use of reverse-field characters in
names is not recommended.
original name = the name that currently appears in the directory
listing. The original name must be spelled exactly as
it appears. Wild cards (? and *) may not be used.
Sending a COPY command using DOS 5.1:
SYNTAX:
>C: backup name=original name
ALTERNATE:
SCO: backup name-original name
EXAMPLES:
>CO: BACKED UP BILLS-BILLS
@C:LETTERS/BU=LOVE LETTERS
71
Sending a COPY command using the command channel:
SYNTAX:
OPEN 15,8,15
PRINT#15, "CO: backup name=original name"
CLOSE 15
ALTERNATE:
OPEN 15,8, 15, "CO:backup name^original name"
CLOSE 15
EXAMPLE #l: (immediate mode)
OPEN 15,8,15
PR I NT# 15, "CO: 60 DAY C0PY=60 DAYS LATE"
CLOSE 15
EXAMPLE #2: (immediate mode)
OPEN 15,8, 15, "CO: TAX FORM COPY=TAX FORM": CLOSE 15
EXAMPLE #3: (program mode)
100 OPEN 15,8,15
■ m • ' • a
'a m m m a
360 PRINT#15, "C:BAME AGAIN=BAME"
■ ■ • ■ •
a a a ■ a
500 CLOSE 15
EXAMPLE #4: (program mode)
100 OPEN 15,8,15
. a - a a a •
a ■ • ■ a
350 F I *= " CO : RECE I PTS . C=RECE I PTS "
360 PRINT#15,FI*
■ a m a m
• • * a •
500 CLOSE 15
EXAMPLE #5:
100 OPEN 15,8,15
(program mode)
■ a • • ■
250 INPUT "NAME OF FILE TO COPY"; ON*
260 INPUT "NEW FILE NAME";NN*
270 PRINT#15,"C:" +NN*+ " = " +0N*
a a a a a
a • a a m
500 CLOSE 15
Examples 4 and 5 illustrate how you can use string variables to store either the entire
disk command (4) or portions of the command (5). You can use this technique in either
immediate or program mode as long as you use the command channel. You cannot
use this technique when using DOS 5.1.
What happens when you issue a COPY command?
72
When you send a COPY command to the disk drive, the disk operating system in the
drive interprets (parses) the command. If the syntax is incorrect, the command is
rejected, the red drive-active light begins flashing and a SYNTAX ERROR report is
generated (error number 30, 31, 32, 33, 34, or 39).
If the syntax is correct, the command is executed. As usual, the red drive-active LED
turns on and the drive motor starts. The head moves directly to track 18 and begins
reading the directory file (18-1, 18-4, and so on). First, it reads through the entire
directory checking for both the original name and the backup name. If the original
name is not found, the command is aborted and the drive reports a 62, FILE NOT
FOUND,00,00 error. If the backup name matches an existing name in the directory,
the command is aborted and the drive reports a 63, FILE EXISTS,00,00 error. If
everything looks OK, the backup name is inserted in the directory and the drive
begins to read the old file and make a duplicate copy. Once the copy is completed
satisfactorily, the diskette's updated Block Availability Map (BAM) and the revised
directory sector are recorded on the diskette, and the drive-active LED and the motor
are switched off.
4.7 Combining Two or More Files
You may use a variation of the COPY command to combine two or more files into a
single, new, larger file. The information will not be merged. The new file will consist
of all the information from old file #1 followed by all the information from file #2, etc.
The command will work with any combination of up to four sequential, program or
user files. It does not work correctly with relative files.
In practice, this command is only used with sequential files. When two program files
are used, the two programs are not linked together properly. If you want to merge
two program segments together, check Section 7.7.
All the files, the new combined file and the old files, must be on the same diskette. If
you need to combine two files that are on separate diskettes, copy them both onto a
single diskette and then combine them. The COPY command that combines two or
more files looks like this:
General -form:
COPYO: combo* ile=0:oldfil el, O:oldfile2
Alternate -form:
C: combo-f i 1 e-ol df i 1 el , ol d-f i 1 e2
Examples:
CO: ALL JUNK=0:OLD JUNK, 0: NEW JUNK
c:all MEMBERS=MALE members, female members
WHERE:
COPY = the disk command. The command COPY may be
abbreviated as C. The zero is the drive number. It
is a holdover from Commodore's dual drives. Its use
is optional on the 1541 disk drive.
73
combofile = the name of the new combined file as it should
appear in the directory. The name may be up to 16
characters long. Any characters except asterisks,
colons, quotation marks, question marks or commas,
may be used in the name. The use of reverse-field
characters in names is not recommended.
oldfilel, oldfile2, = the names of the files to be combined. The names
oldfile3, oldfile4 must be spelled exactly as they appear in the
diskette directory. Wild cards (? and *) may not be
used. The files to be combined do not all have to be
of the same type. The combined file will be the
same file type as the first named oldfile. You may
combine up to four files at one time.
WARNING:
You cannot merge two relative files this way! If you try to use this command to
combine two relative files, the drive will accept the command and begin copying
information into the new combined file. The drive appears to copy all of the infor-
mation from the first relative file correctly. However, part way through the second
file the drive becomes confused and aborts the job with a 66, ILLEGAL TRACK
AND SECTOR,73,01 error. The directory will show the correct name and file type
for the new combined file (you can't have an unclosed REL file), but the length of the
file will be given, as zero blocks. However, the number of blocks free on the diskette
will have decreased indicating that the blocks that make up the new file have been
allocated in the diskette's BAM. The moral? This command just won't work with
relative files.
CAUTION:
The length of the combined file may be wrong! When the DOS creates the directory
entry for the new, combined file, the file length will be set to the sum of the lengths of
the files you have combined. In some cases the actual length will be one block less.
Let's look at an example to see how this can happen. Suppose we have two files,
FILE-A and FILE-B. The directory indicates that FILE-A is three blocks long and
FILE-B is five blocks long. However, a careful examination of what is actually stored
on the diskette using EDIT T&S (Appendix E) reveals that the last block in each file
is not full. FILE-A occupies two full blocks (254 bytes in each) and the first 50 bytes of
the third block. FILE-B occupies four full blocks and the first 132 bytes of the fifth
block. When we combine FILE-A and FILE-B to produce FILE-C, the directory will
indicate that FILE-C is (5+3)= 8 blocks long. However, the FILE-C will actually
occupy only seven blocks; (4+2)= 6 full blocks and the first (50+132)= 182 bytes of the
seventh block. Only when the sum of the number of bytes in the last blocks exceeds
254 will the block count be correct.
Sending a COPY command using DOS 5.1:
SYNTAX:
>C: combo* i 1 e=ol d-f i 1 el , ol d-f i 1 e2
74
alternate:
©CO: combo* i 1 e=0: ol d-f i 1 el , O: ol d-f i 1 e2
EXAMPLES:
>co:all tax=o:fed tax, o: state tax
@c:our stuff=my stuff, your stuff
Sending a COPY command using the command channel:
syntax:
OPEN 15,8,15
PRINT#15, "COrcombof ile=0:oldfilel,0:old-file2"
CLOSE 15
alternate:
OPEN 15,8, 15,"C:combofile=oldfilel,oldfile2"
CLOSE 15
EXAMPLE ttl: (immediate mode)
OPEN 15,8,15
PRINT#15, "C: JUNK=THIS, THAT, THE OTHER"
CLOSE 15
EXAMPLE #2: (immediate mode)
OPEN 15, 8, 15, "CO: 6REEN=0: YELLOW, O: BLUE" : CLOSE 15
EXAMPLE #3: (program mode)
100 OPEN 15,8,15
• • ■ m m
• • ■ a •
360 PR I NT* 1 5 , " C : SPEECH^ I NTRO , BODY , END I NG "
• • • • ■
• • * • *
500 CLOSE 15
EXAMPLE #4: (program mode)
lOO OPEN 15,8,15
• ■ a • •
■ ■ • • ■
350 fi*=»co:expenses=car EXP, HOUSE EXP, FOOD EXP"
360 PRINT#15,FI*
■ ■ a a •
• a • m m m
500 CLOSE 15
EXAMPLE #5: (program mode)
100 OPEN 15,8,15
a • • • ■
• ■ • • m
250 INPUT"NAME OF COMBINED FILE";CF*
260 INPUT"FIRST FILE NAME";N1*
270 INPUT "SECOND FILE NAME";N2*
280 PRINT#15, "C: "+CF*+"="+Nl*+", "+N2*
■ a ■ ■ •
a m m m m
500 CLOSE 15
75
Examples 4 and 5 illustrate how you can use string variables to store either the entire
disk command (4) or portions of the command (5). You can use this technique in either
immediate or program mode as long as you use the command channel. You cannot
use this technique when using DOS 5.1.
What happens when you issue this type of COPY command?
As usual, the disk operating system in the drive interprets (parses) the command. If
the syntax is incorrect, the command is rejected, the red drive-active light begins
flashing and a SYNTAX ERROR report is generated (error number 30, 31, 32, 33,
34, or 39).
If the syntax is correct, the command is executed. The red drive-active LED turns on
and the drive motor starts. The head moves directly to track 18 and begins reading
the directory file (18-1, 18-4, and so on). It reads through the entire directory checking
for both the oldf ile names and the combof ile name. If any of the oldf ile names are not
found, the command is aborted and the drive reports a 62, FILE NOT FOUND,00,00
error. If the combof ile name matches an existing name in the directory, the command
is aborted and the drive reports a 63, FILE EXISTS.00,00 error. If everything looks
OK, the combofile name is inserted into the directory and the drive begins to read the
old files in sequence and copy them into the new combined file. Once the copy is
completed satisfactorily, the updated BAM and the directory sector containing the
combofile name are recorded on the diskette, and the drive-active LED and the motor
are switched off.
4.8 Validating a Diskette
In Commodore jargon, validating a diskette means reading through all the files
stored on a diskette and updating the map that indicates which sectors on the diskette
are free (they are available for use) and which are allocated (they store part of a
program or data file). This map is known as the diskette's Block Availability Map
(BAM). It is stored on track 18, sector 0 of the diskette.
If the Block Availability Map is incorrect, one or more of the files on the diskette may
be overwritten when you store new information on the diskette. When this happens
all of the overwritten information is irretrievably lost. In addition, you will have
crossed files. Two files are crossed when the one or more sectors on the diskette are
part of two different files. You have crossed files if you load, or read, a file and get
what appears to be the first part of one file followed by the last part of another file.
Once you have crossed files on a diskette, the only solution is to copy the files onto
another diskette and reformat the original.
The most common causes of errors in a diskette's BAM are scratching improperly
closed files and using the save-and-replace command on an almost full diskette. You
can prevent crossed files and other horrors by validating your working diskettes
regularly. Read-only diskettes such as commercial programs do not need to be
validated.
A VALIDATE command looks like this:
76
General -form:
VALIDATE©
Alternate form:
V
Examples:
VO
V
WHERE:
VALIDATE = the disk command. The command VALIDATE may be
abbreviated as V. The zero is the drive number. It is a
holdover from Commodore's dual drives. Its use is
optional on the 1541 disk drive.
CAUTIONS:
1. Do nof validate commercial games diskettes. Many commercial games are supplied
on diskettes that have elaborate protection schemes to prevent you from making a
backup copy of the diskette. Sometimes the diskette's BAM has been deliberately
altered as part of the protection. If you remove the write protect tab and validate
the diskette, the game may never work again. Don't do it!
2. Do not validate diskettes containing random access files. Some programs store
data on a diskette in random access files. Note that a random access file is not a
relative file. If you validate a diskette that contains a random access file, you may
lose all the data stored in the file!
3. Validation may free deliberately allocated blocks. The CHECK DISK program on
your 1541TE ST/DEMO diskette scans a disk for bad sectors. If it finds one, that
sector is deliberately allocated in the BAM so that the drive will not attempt to
store any information in that sector. When you validate the diskette, that bad
sector will be freed. When the drive attempts to store information there, you will
get a disk error. If you use CHECK DISK to allow you to keep using a flawed
diskette, you must rerun CHECK DISK after you validate the diskette. It is a
much better idea to junk any diskette that contains a flaw!
Sending a VALIDATE command using DOS 5.1:
SYNTAX:
>V
ALTERNATE:
evo
EXAMPLES:
>VO
@V
77
Sending a VALIDATE command using the command channel:
SYNTAX:
OPEN 15,8,15
PRINT#15, "V"
CLOSE 15
ALTERNATE:
OPEN 15,8, 15," VO": CLOSE 15
EXAMPLE #1: (immediate mode)
OPEN 15,8,15
PRINT#15, "VO"
CLOSE 15
EXAMPLE #2: (immediate mode)
OPEN 15,8, 15, "VO": CLOSE 15
EXAMPLE #3: (program mode)
100 OPEN 15,8,15
■ ■ ■ • •
360 PRINT#15, "V"
• * ■ • a
■ • « « m
500 CLOSE 15
EXAMPLE #4: (program mode)
100 OPEN 15,8,15
• ■■■«
• m • m m
350 FI*="VO"
360 PRINT#15,FI*
• m m m •
■ a ■ • •
500 CLOSE 15
Example 4 illustrates how you can use a string variable to store the disk command.
You can use this technique in either immediate or program mode as long as you use
the command channel. You cannot use this technique when using DOS 5.1.
What happens when you issue a VALIDATE command?
When you send a VALIDATE command to the disk drive, the disk operating system
in the drive interprets (parses) the command. If the syntax is incorrect, the command
is rejected, the red drive-active light begins flashing and a SYNTAX ERROR report
is generated (error number 30, 31, 32, 33, 34, or 39).
If the syntax is correct, validation begins. As always, the red drive-active LED turns
on and the drive motor starts. While the drive is getting up to speed, the disk operating
system creates a new, blank Block Availability Map (BAM) in the disk drive's RAM
memory. In this new map all sectors on the directory are free (available for use). Once
the drive is up to speed, the head moves to track 18 and reads the first directory entry.
This file is traced across the diskette by following the track and sector pointers. Each
78
of the sectors (blocks) used to store the file is marked as in use (allocated) in the BAM.
Each file in the directory is traced. Once all files in the directory have been traced,
the revised BAM is recorded on track 18, sector 0 of the diskette and validation is
complete.
If an improperly closed file is encountered in the directory, the file is not traced. All
that happens is that the file type byte in the directory entry is changed to zero. This
makes the file a scratched file. It will no longer appear in the directory. Since the file
is not traced, any blocks that were used to store the file are freed in the BAM and
become available for storage of new programs or data. Note that this is different from
what happens when you SCRATCH an improperly closed file. In both cases the file is
converted into a scratched file. However, with the SCRATCH command the sectors
that had been allocated to the unclosed files are not freed in the BAM. These blocks
can only be freed by use of the VALIDATE command. To avoid problems with the
BAM, never scratch an improperly closed file.
NOTES:
If a disk error is encountered while reading any file, the entire validation process
stops. No harm has been done because the diskette's original BAM is still intact.
However, you can't validate the diskette either. The only solution is to copy the intact
files one by one onto another diskette and then do a full NEW on the problem diskette
to eliminate the errors.
Appendix E contains two programs that help you spot and localize problems on a
diskette. The FULL DIRECTORY program allows you to do a quick check for BAM
errors. It will point out any inconsistencies between the number of blocks free in the
directory and in the BAM. The CONFIRM ALL FILES program does a much more
comprehensive job of checking. It first checks the internal consistency of the BAM
and then traces all files. Any problems, such as crossed files or inconsistencies between
the actual length of the files and the length shown in the directory, are displayed. Full
instructions for using these programs are given in Section 11.1.
4.9 Your BASIC Housekeeper
Doing diskette housekeeping is something like doing everyday household chores.
Nobody really likes doing them, but they have to be done. To make your life a little
more pleasant, let me introduce you to your BASIC housekeeper.
The BASIC housekeeper is a menu-driven BASIC program that makes doing diskette
housekeeping a lot easier. When you RUN the program, you will be presented with a
list of the housekeeping tasks and asked which you would like to do. After selecting
the task you want, you will be prompted for all the necessary information to carry out
the task. It should make your housekeeping chores less difficult.
*******************
PROGRAM: HOUSE HELP
*******************
0 REM 1541 USER'S GUIDE SECTION 4.9
1 REM COPYRIGHT: G. NEUFELD, 19G4
2 :
79
3 REM 1541 DISKETTE HOUSEKEEPING
4 :
1 OO P0KE53280 ,15: P0KE5328 1,15
1 lO PRINTCHR* ( 142) CHR* (8) CHR* < 144)
120 Z*=CHR*tO)
1 30 PR I NTCHR* < 1 42 ) CHR* ( 8 ) CHR* (144)
140 D*=" {HOME} {DOWN 22}"
150 DIM F*(144)
160 OPEN 15,8,15
170 PRINT" {CLR} 1541 DISKETTE HOUSEKEEPI
NG"
i8o print" <:down>{rvs>c<:roff>opy A FILE"
190 PRINT" {DOWN} {RVS}D{ROFF} I SPLAY DIREC
TORY"
200 PRINT" {DOWN} {RVS} I {ROFF}NITIALIZE DI
SKETTE"
2io print" <:down><:rvs>n<:roff>ew a diskett
E"
220 PRINT" {DOWN} {RVS}R{ROFF}ENAME A FILE
ii
230 PRINT" {DOWN} {RVS}S{ROFF}CRATCH A FIL
E"
240 PRINT" {DOWNXRVSJVtROFFJAL I DATE A DI
SKETTE"
250 print" <:down><:rvs}t<:roff>erminate PRO
GRAM"
260 PRINT" CDOWN> WHICH OPTION? ";
270 GET A*: IF A*="" GOTO 270
280 PR I NT A*
290 IF A*="C" GOTO 410
300 IF A*="D" GOTO 680
310 IF A*="I" THEN PRINT#15, "IO":DF=0:GO
TO 170
320 IF A*="N" GOTO 890
330 IF A*="R" GOTO 1470
340 IF A*="S" GOTO 1690
350 IF A*="T" THEN CLOSE 15: PRINT" {CLR}"
:END
360 IF A*="V" GOTO 2230
370 GOTO 170
380 :
390 REM COPY A FILE
400 :
410 GOSUB 2380:IF FLOO GOTO 170
420 DF=0
430 PRINT" {CLR} COPY FILE"
440 PR I NT "{DOWN} YOU CAN MAKE A BACKUP CO
PY OF ANY PRG, "
450 PRINT"SEQ, OR USR FILE ON THE SAME D
ISK. "
80
460 PR I NT "YOU CANNOT MAKE A COPY OF A RE
L FILE"
470 PR I NT "OR COPY A FILE TO ANOTHER DISK
ETTE. "
480 PR I NT "USE COPY /ALL FROM THE 1541 TEST
/DEMO"
490 PRINT "DISKETTE FOR THESE APPLICATION
S. "
500 INPUT" {DOWN} OK TO CONTINUE (Y/N) YC
LEFT 3>";A*
510 IF (ASC(A*)AND127)<>89 GOTO 170
520 PRINT" {DOWN} FILE TO COPY (WILD CARDS
ARE OK>"
530 F*= " " : I NPUT " F I LE NAME " ; F*
540 IF F*="" GOTO 430
550 GOSUB 2510: IF FL=1 G0T0430
560 PRINT" {DOWN}FOUND FILE NAMED: "F*"
II
570 INPUT" {DOWN* IS THIS THE CORRECT FILE
(Y/N) Y{LEFT 3} "5 A*
580 IF (ASC (A*) AND 127)089 GOTO 430
590 BF*=LEFT*(F*, 13)+"/BU"
600 PRINT" {DOWN} BACKUP COPY NAME: "BF*
610 PRINT" {UP} "TAB (17);: INPUT BF*
620 PR I NT# 1 5 , " C : " +BF*+ " - " +F*
630 GOSUB 2420
640 GOTO 170
650 :
660 REM DISPLAY THE DIRECTORY
670 :
680 PRINT" {CLR} DISPLAY DISKETTE DIRECTOR
Y"
690 IF DF=1 GOTO 730
700 PRINT" {DOWN} ONE MOMENT PLEASE "
71 O GOSUB 2720:IF FL=1 GOTO 170
720 DF=1
730 T=0: PRINT" {CLR} DISKETTE DIRECTORY
{DOWN} "
740 FOR K=l TO NE
750 PRINTTAB(T)F*(K)
760 IF INT(K/20)OK/20 GOTO 820
770 IF T=0 THEN PRINT" {HOME} {DOWN}": T=20
:GOTO 820
780 PRINT D* "{DOWN}PRESS {RVS} RETURN {RO
FF} FOR MORE"
790 GET A*: IF A*<>"" GOTO 790
800 GET A*: IF A*OCHR*(13) GOTO 800
810 T=0: PRINT" {CLR} DISKETTE DIRECTORY
{DOWN} "
820 NEXT K
81
830 PRINT D* "{DOWN3 PRESS {RVS* RETURN £R0
FFJ FOR MENU"
840 GET A*: IF A*<>"" SOTO 840
850 GET A*: IF A*OCHR*<13> GOTO 850
860 GOTO 170
870 :
880 REM NEW A DISKETTE
890 TRY=0:DF=0
900 PRINT" {CLRJ FORMAT A DISKETTE"
910 PRINT" {DOWN J ALLOWS YOU TO FORMAT A D
ISKETTE"
920 PR I NT "{DOWN* FULL NEW = REFORMAT ENT
IRE DISKETTE"
930 PRINT" {DOWN> SHORT NEW = ERASE OLD D
I RECTORY & BAM"
940 INPUT" {DOWN3 OK TO CONTINUE (Y/N) Y£
LEFT 33-"; A*
950 IF <ASC(A*>AND127)<>89 GOTO 170
960 PRINT" <UP> INSERT DISKETTE & PRESS {R
VS>RETURN{ROFF> "
970 GET A*: IF A*O""60T0 970
980 GET A*: IF A*OCHR* ( 1 3 ) GOTO 980
990 PRINT#15„ "M-W"CHR* (6) CHR* (O) CHR* (2) C
HR*(18)CHR*<0)
1 000 PR I NT# 15," M-W " CHR* < O ) CHR* < O ) CHR* < I )
CHR* (176)
101O PRINT#15, "M-R"CHR*<0)CHR*(0)
1020 GET#15,A*:A=ASC(A*+Z*)
1030 IF AM27 GOTO 1010
1040 TRY=TRY+1
1050 IF AOl AND TRY<5 GOTO 990
1060 IF A>1 THEN FL=1:G0T0 1320
1070 PRINT#15, "M-R"CHR* (22) CHR* (O) CHR* <2
)
1080 GET#15,A*:IF A*=""THEN A*=Z*
1090 ID*=A*
1100 GET#15,A*:IF A*=""THEN A*=Z*
1110 ID*=ID*+A*
1120 OPENl,8,4,"#"
1130 PRINT#15, "Ul:4 O 18 O"
1140 PRINT#15,"B-P:4 144"
1150 DN*="":FbR K=l TO 24
1160 GET#1,A*:IF A*=""THEN A*=Z*
1170 DN*=DN*+A*
1180 NEXT
1190 CLOSE 1
1200 PRINT" {DOWN* THIS DISKETTE IS ALREAD
Y FORMATTED"
1210 PRINT" {DOWN J DISK NAME: "LEFT*(DN*, 1
6)
82
1220 PR I NT "COSMETIC ID: "MID* (DN*, 19,2)
1230 PR I NT "EMBEDDED ID: "CHR* (34)CHR*(20
) ID*
1240 INPUT" {DOWN* FULL NEW, SHORT NEW, OR
QUIT(F/S/Q)";A*
1250 A*=LEFT* (A*, 1 )
1260 IF A*="F" THEN FL=1:G0T0 1320
1270 IF A*="S" THEN FL=0:GOTO 1320
1280 IF A*="Q" GOTO 170
1290 PRINT" CDOWN}ENTER ONLY *F', *S*, OR
*Q*"
1300 GOTO 1240
1310 :
1320 INPUT" {DOWN} NEW DISK NAME:";DN*
1330 IF FL=1 THEN ID*="": INPUT" <: DOWN >FOR
MATTING ID"; ID*
1340 IF LEN(ID*)<>2 THEN PRINT" {DOWN} ID
MUST BE 2 CHARACTERS LONG": GOTO 1330
1350 IF FL=0 THEN PRINT" <DOWN>CAUTION: A
LL FILES WILL BE LOST"
1360 IF FL=1 THEN PRINT" {DOWN} CAUTION: A
NY OLD DATA WILL BE LOST"
1370 INPUT" {DOWN J OK TO CONTINUE (Y/N) N
CLEFT 3>";A*
1380 IF <ASC(A*)AND127)<>89 GOTO 170
1390 PRINT" { DOWN > WORKING "
1400 IF FL=1 THEN PRINT#15, "N: "+DN*+" . "+
ID*
1410 IF FL=0 THEN PRINT#15, "N: "+DN*
1420 GOSUB 2420:NE=0
1430 GOTO 170
1440 :
1450 REM RENAME A FILE
1460 :
1470 GOSUB 2380:IF FLOO GOTO 170
1480 DF=0
1490 PRINT" CCLRJ RENAME A FILE"
1500 PRINT" {DOWN} ALLOWS YOU TO CHANGE TH
E NAME OF ANY"
1510 PRINT"FILE ON THE DISKETTE"
1520 INPUT" {DOWN} OK TO CONTINUE (Y/N) Y
CLEFT 3>";A*
1530 IF (ASC(A*)AND127)<>89 GOTO 170
1540 PRINT" CDOWN>FILE TO RENAME (WILD CA
RDS ARE OK)"
1 550 F*= " " : I NPUT " F I LE NAME " 5 F*
1560 IF F*="" GOTO 1490
1570 GOSUB 2510: IF FL=1 GOTO 1490
1580 PRINT" CDOWN> FOUND FILE NAMED: "F*"
83
1590 INPUT" {D0WN> IS THIS THE CORRECT FIL
E <Y/N) YO.EFT 3>";A*
1600 IF <ASC(A*)AND127)<>89 GOTO 1490
1610 PRINT" {DOWN > NEW FILE NAME: "F*
1620 PRINT" OJP> "TAB < 14) ;: INPUT NF*
1630 PRINT#15, "R: "+NF*+"="+F*
1640 GOSUB 2420
1650 GOTO 170
1660 :
1670 REM SCRATCH A FILE
1680 :
1690 PRINT" iCLRy SCRATCH A FILE"
1700 PRINT" < DOWN 3 ALLOWS YOU DELETE A FIL
E FROM THE"
1710 PRINT "DISKETTE'S DIRECTORY. YOU MAY
NOT"
1720 PR I NT "SCRATCH AN UNCLOSED FILE. USE
THE"
1730 PR I NT "VALIDATE OPTION FOR THESE."
1740 INPUT" <DOWN>OK TO CONTINUE <Y/N) Y
{LEFT 3>";A*
1750 IF <ASC<A*)AND127)<>89 GOTO 170
1760 PRINT" <DOWN>FILE TO SCRATCH (WILD C
ARDS ARE OK) "
1770 F*="": INPUT" FILE NAME";F*
1780 IF F*="" GOTO 1690
1790 FOR K=LEN(F«) TO 1 STEP -1
1800 C*=MID*(F*,K, 1)
1810 IF C*="*" OR C*="?" GOTO 1920: REM W
ILD CARDS
1820 NEXT
1830 GOSUB 2510: IF FL=1 GOTO 1690
1840 PRINT" <DOWN> FOUND FILE NAMED: "F*"
II
1850 INPUT" {DOWN>SCRATCH THIS FILE (Y/N)
Y CLEFT 3>";A*
1860 IF (ASC<A*)AND127)<>89 GOTO 1690
1870 PRINT#15, "S: "+F*
1880 DF=0
1890 GOSUB 2420
1900 GOTO 170
1910 :
1920 K=l:NEXT
1930 IF DF=1 GOTO 1970
1940 PRINT" {DOWN 3 ONE MOMENT PLEASE "
1950 GOSUB 2720:IF FL=1 GOTO 170
1960 DF=1
1970 PRINT"tDOWNJTHE FOLLOWING FILES WIL
L BE SCRATCHED {DOWN> "
1980 NS=0:LN=LEN(F*):FOR K=l TO NE:FD=0
84
1990 FOR C=l TO LN
2000 X*=MID*(F*,C,1):IF X*="?"GOTO 2030
2010 IF X*="*"THEN C=LN:FD=l:G0T02030
2020 IF X*<>MID*(F*(K),C,,1) THEN C=LN:FD
=-1
2030 NEXT C
2040 IF FD=-1 GOTO 2060
2050 PR I NTF* ( K ) : NS=NS+ 1 : ND ( NS ) =K
2060 NEXT K
2070 IF NS=0 THEN PRINT "NO MATCHING NAM
ES FOUND": GOTO 2160
2080 PRINT "CDOWN}"NS" MATCHING NAMES FOU
ND"
2090 INPUT" CDOWN>OK TO CONTINUE (Y/N) Y
CLEFT 3>";A*
2100 IF (ASC(A*)AND127)<>89 GOTO 170
2110 PRINT#15, "S:"+F*
2120 FOR K=l TO NS
2130 F* <ND (K) ) =" CRVS> "+F* (ND <K> )
2140 NEXT
2150 GOSUB 2420
2160 PRINT" CDOWN}PRESS CRVS5 RETURN CROFF}
FOR MENU"
2170 GET A*: IF A*<>"" GOTO 2170
2180 GET A*: IF A*OCHR*<13) GOTO 2180
2190 GOTO 170
2200 :
22 lO REM VALIDATE A DISKETTE
2220 :
2230 GOSUB 2380: IF FLOO GOTO 170
2240 PRINT" <CLR> VALIDATE A DISKETTE"
2250 PRINT" {DOWN* CREATES A NEW BAM THAT
MATCHES THE"
2260 PR INT "FILES LISTED IN THE DIRECTORY
■
2270 PRINT" IN ADDITION, ANY UNCLOSED FIL
ES IN THE"
2280 PR I NT "DIRECTORY WILL BE ELIMINATED.
II
2290 INPUT" {DOWN} OK TO CONTINUE (Y/N) Y
CLEFT 3}"; A*
2300 IF (ASC ( A* > AND 127)089 GOTO 170
2310 PRINT" {DOWN} THIS CAN TAKE UP TO 2 M
INUTES. . . "
2320 PRINT#15, "VO"
2330 GOSUB 2420
2340 GOTO 170
2350 :
2360 REM I NIT DISK & READ ERROR
2370 :
85
2380 PRINT#15, "IO"
2390 :
2400 REM READ ERROR STATUS
2410 :
2420 INPUT#15«EN,EM*,ET,ES
2430 IF EN<20 THEN FL=0: RETURN
2440 PRINT" {DOWNMRVSJDISK ERROR {DOWN}"
2450 PRINT" "ENSEM*;ET;ES
2460 PRINT "{DOWN} PRESS {RVS}RETURN<ROFF}
TO CONTINUE"
2470 GET A*: IF A*<>""GOTO 2470
2480 GET A*: IF A*OCHR*<13> GOTO 2480
2490 FL=l: RETURN
2500 :
2510 OPEN 1,8,4, "#"
2520 OPEN2,8,5,F*
2530 GOSUB 2420: IF FL=1 THEN CLOSE l:CLOS
E2: RETURN
2540 PRINT#15, "M-R"CHR*(144> CHR*<2>
2550 GET#15,A*:SE=ASC(A*+Z*>
2560 PR I NT# 1 5 , " M-R " CHR* < 1 48 ) CHR* < 2 )
2570 GET#15,A*:PT=ASC(A*+Z*>
2580 PRINT#15, "Ul:4 O 18"SE
2590 PR I NT# 15," B-P : 4 " PT+3
2600 F*=""
2610 FOR K=1T016:GET#1,A*:IFA*=""THENA*=
Z*
2620 F*=F*+A*:NEXT
2630 FOR K=16T01STEP-1
2640 IF ASC<MID*(F*,K)>0160 GOTO 2670
2650 NEXT
2660 F*=LEFT*(F*,K):GOTO 2690
2670 F*=LEFT*(F*,K):K=l:NEXT
2680 CLOSEl:CLOSE2
2690 FL=0: RETURN
2700 :
2710 REM READ THE DIRECTORY
2720 se=i:ne=o
2730 0PEN1,8,4,"#"
2740 PRINT#15, "Ul:4 O 18"SE
2750 GOSUB 2420: IF FL=1 THEN CL0SE1:RETU
RN
2760 FOR K=0 TO 7
2770 PR I NT# 1 5 , " B-P : 4 " 2+K*32
2780 GET#1,A*:FT=ASC(A*+Z*)
2790 GET#1,A*:TL=ASC<A*+Z*>
2800 GET#1,A*:SL=ASC(A*+Z*)
2810 IF SL=0 AND TL=0 GOTO 2940
2820 NE=NE+1
2830 F*<NE>=""
86
2840 IF FT=0 THEN F*(NE)=,,{RVS>"
2850 IF FT>0 AND FT<128 THEN F*(NE)="€RV
S>*{R0FF> "
2860 FOR L=1T016
2870 GET#1 , A*: IFA*=" "THENA*=Z*
2880 F*<NE>=F*(NE)+A*
2890 NEXT L
2900 FOR P=LEN<F*(NE>> TO 1 STEP-1
2910 IF ASC<MID*(F*(NE),P)X>160 GOTO 29
30
2920 NEXT
2930 F* <NE) =LEFT* (F* (NE) , P)
2940 NEXT K
2950 PRINT#15, "B-P:4"0
2960 GET# 1 , A* : TL=ASC ( A*+Z* )
2970 GET#1,A*:SE=ASC(A*+Z*>
2980 IF TL=18 GOTO 2740
2990 CLOSE 1
3000 FL=0: RETURN
87
CHAPTER
5
INTRODUCTION TO FILE HANDLING
This chapter is designed to introduce you to the main ideas and commands related to
file handling. Each section includes several examples to help you learn how to make
use of the excellent file-handling capabilities of your Commodore disk drive. You
should read through this chapter before you read the chapters that deal with specific
types of files.
5.1 Separating Data from a Program
Most people purchase a disk drive to enable them to load or save programs quickly
and easily. However, the real power of your disk drive is not its speed in loading or
saving programs. The real power lies in its ability to store and retrieve large amounts
of information (data), independent of the program that produces or uses this
information.
A program is a set of instructions to tell the computer how to do a task, such as how to
set up and maintain a mailing list. The actual information, data, used by a program
may or may not be part of the program. If it is part of the program, the information is
usually stored in DATA statements. However, data is usually stored separately in a
data file. The ability to separate the program (how to do a task) from the data (the
information needed for a particular task) is very important. It allows you to write
powerful, general purpose programs.
Most commercial programs store the data used or produced by the program in a data
file stored on a diskette. Here are a few examples of types of programs that use this
technique:
Program File Data File Produced
Mailing list program File of members' names & addresses
Grade book program File of students' names & grades
Word processor File of letters, documents, etc.
Accounting program File of debit/credit transactions
Once the information used by the program is separated from the program itself, the
program usually becomes more useful. A grade book program is a good example.
Suppose you wrote a grade book program with the students' names and marks in
DATA statements within the program. You would end up making many copies of the
program, one for each class, subject or grade level you taught. If you had stored the
89
names and marks in data files, you would need only one copy of the program. The
same program would handle all your different classes, it would just access different
data files.
5.2 What a Data File Is
A data file is a big, long string of information. We will use the file of information
produced by a mailing list program to illustrate the structure of a data file. Here is
the start of our data file.
NOTE:
* represents a carriage return character, CHR$(13)
Lee*Jim»15 Fifth St.»Anytown«Anystate*123472*SiiiithBon»Sally»4-S67 Second Ave
^
•>
Anytown»Anystate*l 25476«Jones«Ti m»23 First St.»Anytown»Anygtate»12345&»
The information in a data file is divided up into clumps called records. Each record
contains information about one particular person, place or thing. Records are usually
separated from each other by a carriage return character, CHR$(13). One record in
our mailing list file would probably contain the information about one member —his
or her name, address and zip code.
The first part of the file (composed of three records):
Jim's Record
Sally's Record
Tim's Record
Within each record, the information is often divided into fields. Each field contains a
specific piece of information about one person, place or thing. The fields within a
record are usually separated from each other by a special character called a delimiter
or separator. A carrige return character, CHR$(13), or a comma is often used as a
delimiter.
Record #1 (45 characters grouped into seven fields):
NOTE:
* represents a carriage return character, CHR$(13)
Lee*Jim*15 Fifth St. ♦Anytown*Anystate» 123472*
Record #2 (57 characters grouped into seven fields):
90
Smithson*Saily*4-567 Second Ave.*Anytown«Anystate*123476*
Record #3 (47 characters grouped into seven fields):
Jones»Tira#23 First St. »Any town* Any state* 123456*
Note that in this data file the records are not the same length. Record #2 is longer
than either of the other records. This is simply because Sally Smithson's name and
address contain more characters.
Since we will be using the terms file, record, field, and delimiter throughout this
section, let's go over them one more time.
File: Contains information about all members of the group (e.g., information about
Jim Lee, Sally Smith, etc.).
Record: Contains information about one member of the group (e.g., all information
about Jim Lee).
Field: Contains one piece of information about one member; separated from other
fields with a delimiter (e.g., Jim Lee's address: 15 Fifth Steet).
Delimiter: A special character that is used between records or fields as a separator.
There are four characters that you can use as delimiters:
Delimiter Character ASCII For Use Between
a null character CHR$(0) not recommended
a carriage return CHR$(13) records or fields
a comma CHR$(44) fields only
a colon CHR$(57) fields only
5.3 Kinds of Files
There are four main types of files that you can use on your 1541 disk drive. They are
program files, sequential files, relative files and user files. The file type abbreviations
that appear in a directory listing and the main uses of these major types of files are
summarized below:
File Type Directory Main Uses
Program files PRG Storing copies of RAM memory
such as programs, routines and
some data.
Sequential files SEQ Storing limited amounts of data or
data that is rarely modified.
91
Relative files REL Storing large amounts of data that
needs frequent updating.
User files USR User designed files. For ad-
vanced programmers and special
tasks only.
Unclosed Files
Occasionally you may come across a file that has an * in front of its file type in the
directory (e.g, *PRG, *SEQ, *REL or *USR). This is a file that has not been closed
properly. This happens when you forget to close a file after you have written to it. Be
sure to read the warning about unclosed files.
WARNING:
Do not scratch an unclosed file! If you want to eliminate an unclosed file from the
directory, validate the diskette (see Section 4.8). If the data is very important, you can
try to recover the file (see Section 11.3).
Locked Files
Occasionally you may run across a file that has a < following its file type in the
directory (e.g., PRGK SEQ<, RELK or USR<). This is a locked file. It cannot be
scratched from the directory unless you first unlock it. You cannot lock or unlock a
file easily. You have to change the file type byte in the diskette's directory using
Block-Read (Ul:) and Block-Write (U2:) commands. The MOD ENTRY program
given in Section 13.8 makes this easy to do.
Unusual File types
Once in a while you may encounter a file with a very odd looking file type (e.g, DEL,
SR? or EL?). These strange file types are not very useful. They are used more for
shock value than anything else. These unusual file types are produced by changing
the file type byte in the directory entry to an abnormal value. The way to do this is
described in Section 13.8.
5.4 Accessing a File
Using a data file on disk is very similar to using an ordinary filing cabinet. If you
want to store or retrieve information in a filing cabinet you follow a three step
procedure:
1. Open the drawer of the filing cabinet
2. Remove or insert some information.
3. Close the drawer again.
To use a data file on disk, you follow a similar procedure:
1. Open the data file for access using an OPEN statement.
2. Read or write some information in the file.
3. Close the data file using a CLOSE statement.
92
CAUTION:
As with any filing system, careful attention to detail is important. If your filing habits
are sloppy, you will misfile or even lose information.
Let's take a look at each of the steps in using a data file more closely.
5.5 Opening a File
Opening a data file is really establishing a data communication link between your
program and some device. Your computer is a very sophisticated communications
device that can communicate with many different devices in a variety of ways. As a
result, we have to specify a number of different things when we establish our link:
1. A logical file number to identify this particular data communication link. This is
necessary because several links can be in use at one time.
2. A device number to specify the device with which we want to communicate. This
is necessary because your computer can communicate with a variety of devices
(e.g., disk drive=8, tape recorder=l, screen=3 and printer=4).
3. A channel number to specify the nature of the communication. Its meaning
depends on which device you are using. It is also called the secondary address.
4. File identification information to identify the specific file we wish to access and
whether we want to read or write information. This is not necessary for devices
that do not contain named files such as printers and keyboards.
All this information is specified in a single OPEN statement. Since there is a lot of
information to be specified, let's go over the structure of an OPEN statement carefully.
NOTE:
Since this book deals with using your disk drive, we won't go into all the possible
variations of the OPEN statement. Please consult your user's manual when using
other devices.
SYNTAX:
OPEN i i 1 e#, devi ce#, channel #, "fid"
EXAMPLES:
OPEN 15,8,15
OPEN 1,8,5, "HOME ACCOONTS,SEQ, WRITE"
OPEN 6,8,3, "FAST COPY,PRG,R"
OPEN 2,8, 12, "ARTICLES, L, "+CHR* ( 169)
WHERE:
file# = logical file number (integer 1-127).
device# = device number (eight for the disk drive).
93
channel# = channel number (2-14 are data channels, 15 is the command
channel).
fid = file identification information (required for data files,
optional for the command channel).
Before we examine each part in detail here are two typical OPEN statements:
ANNOTATED EXAMPLES:
OPEN 15,8,15
Opens file# 15 to the disk drive's command channel.
OPEN 3,8,4, "O: BUNCHADATA, S,R"
Opens file# 3 to device eight (drive) using channel* 4 to access the file on drive zero
named BUNCHADATA. It is a sequential file (S) and we want to read (R) it.
5.6 The Logical File Number (file#)
Your VIC-20 or Commodore 64 allows you to have several files open at once. For
example, at some point in your program you might be reading data from one disk file,
writing data to another disk file and sending information to a printer. As a result,
there has to be some simple way for you to indicate which file you want to access. This
is the function of the logical file number.
When you establish the communications channel using an OPEN statement, the file
you want is assigned the logical file number that you specified. Until you close down
the channel, you refer to it by its logical file number. For example, the logical file
number of this file is five.
OPEN 5,8,7," BUNCHADATA , S , W "
PRINT#5,"HI MOM"
CLOSE 5
The OPEN statement establishes the association between the disk file, BUNCH-
ADATA, and the logical file number five. Subsequent statements that refer to this
file, refer to it by its logical file number five.
NOTES:
1. Use integers between one and 127 as logical file numbers. You can use any integer
between one and 255 as a logical file number. However, the integers greater than
127 act differently. These numbers cause your computer to send a line feed
character, CHR$(10), as well as a carriage return character, CHR$(13), at the end
of each line. This wastes disk space and makes it impossible for you to read the file
using an INPUTS statement.
2. You can use different numbers for the same file. Many people are under the
mistaken impression that you have to use the same file number to read a data file
as you used to write the file. This is not true. Logical file numbers are only
94
temporarily associated with a particular file. Once a file has been closed, there is
no longer any association between a number and a file.
3. All files open must have different logical file numbers. Since the logical file
number is used to identify one particular file, you must use different logical file
numbers for each of the files open simultaneously. Once you close a file, you are
free to reuse that logical file number in your program.
5.7 The Device Number (device-*)
This number specifies the device with which you want to communicate. Your 1541
disk drive is normally device number eight. As long as you are only using one drive,
always use this number and ignore the rest of this section.
If you want to use two 1540 or 1541 disk drives simultaneously you will have to
change the device number of one of the drives. You can change the device number of
your disk drive either temporarily or permanently.
Permanently Changing a 1541s Device Number
If you own two or more disk drives, you may want to permanently change the device
numbers on all but one of the disk drives. This way you will avoid the bother of
changing the device numbers every time you fire up your system. Section 12.3
describes how to make this modification.
Temporarily Changing a 1541's Device Number
The procedure outlined below is useful when you have temporarily borrowed a second
or third drive. The change of device number is temporary. When you turn the drive
OFF and then ON again, the device number will be reset to eight.
1. Turn ON your computer and all your disk drives. Connect the drive, whose device
number is to be changed, to the serial bus on the computer. Disconnect the other
drive(s) from the serial bus (otherwise you will change the device number of all
the drives).
2. Execute the following commands from either immediate mode or from within a
program:
OPEN 15,8,15 (open the command channel)
PRINT#15, "M-W"CHR* ( 1 19) CHR* (O) CHR* <2)
CHR* (dn+32) CHR* (dn+64)
CLOSE 15
WHERE:
dn = the device number desired (an integer 4-15; values greater than
15 may cause problems).
For example, to change a 1541 to device number nine, you would use:
95
OPEN 15,8,15 (open the command channel)
PRINT#15, "M-WCHR* ( 1 19) CHR* <0) CHR* <2)
CHR*<41)CHR*<73)
CLOSE 15
3. Connect one more disk drive to the serial bus. This newly attached drive will be
device* 8. If you are only using two drives, you are now finished. If you are using
three or more drives, repeat Steps 2 and 3 for each of the drives (be sure to give
them all different device numbers). Note that you do not have to disconnect a drive
once its device number has been changed. Since it is no longer device number
eight, it will ignore the subsequent commands.
WARNING:
Be sure you use the commands exactly as indicated above. The format and examples
given on page 40 of the 1541 User's Manual are incorrect. The correct syntax for a
Memory-Write command is "M-W" not "M-W:". If you include the colon, the 1541's
device number will notbe changed!
HINT:
If you find it inconvenient to unplug the disk drives from the serial bus, you can turn
OFF a disk drive instead of disconnecting it as indicated in Step 1. Once you have
completed Step 2, turn the disk drive ON again.
5.8 The Channel Number (channel#)
The channel number is also known as the secondary address. It specifies the nature of
the communication desired. Its specific meaning depends on the device with which
you are communicating. Because its meaning varies so much, we will only consider
its meaning as it relates to the disk drive.
Channel* Normal Use
0 Reserved by DOS for loading programs.
1 Reserved by DOS for saving programs.
2-14 Read or write to any type of disk file.
15 Disk drive command channel.
16-17 Reserved by DOS for internal use.
18+ Illegal values.
WARNING:
Every file that is open musthave its own channel*. Once you have closed a file, you
are free to reuse that channel*.
The Command Channel
The 1541 disk drive can carry out a number of disk operations, such as formatting a
disk or scratching a file, without direct intervention by the computer. The command
channel is used to send commands to the disk drive when you want it to carry out
these types of operations. The housekeeping commands are described in detail m
Chapter 4. Appendix D contains a summary of all the diskette-housekeeping and
direct-access commands. The command channel is also used by the disk drive to
report disk errors.
Unusual Uses of Channel Numbers
Channel# 0 is normally reserved for use by the computer for loading programs. If you
use a channel number of zero when opening a file to read it, you will get normal
results unless you are trying to access the directory file, $. In this case, the DOS will
convert the directory file into a pseudo program in the same way it does when you
LOAD the directory. This is sometimes a useful approach to use when reading a
diskette's directory from within a program. If you attempt to use a channel number of
zero when opening a file in write mode, the file will not be opened and the drive will
report a 62, FILE NOT FOUND, error message.
Channel# 1 is normally reserved for use by the computer for saving programs. If you
use a channel number of one when opening a file to write to it, you will get normal
results. If you attempt to use a channel number of one when opening a file in read
mode, the file will not be opened and the drive will report a 63, FILE EXISTS, error
message. The drive apparently overrides your read mode declaration and is attempt-
ing to prevent you from overwriting an existing file.
5.9 File Identification Information
The last part of an OPEN statement is used to identify the specific file you want to
access.
OPEN file#,device#,channel#, "rFid"
Program, Sequential and User Files
The "fid" syntax for program, sequential, and user files is:
SYNTAX:
"dri ve#:-f i lename, type, mode" (for new or existing files)
" dr i ve# : -f i 1 ename " (for existing files only)
EXAMPLES:
"0:BUNCHA STUFF, SEQ, READ"
" O : ACCOUNTS , USR , W "
"FAST COPY, PRG, WRITE"
"BIGFILE"
WHERE:
drive# = the drive number (always zero on the 1541); optional on all
Commodore single disk drives.
filename = the name of the file as given in the directory; may contain
wild cards if file already exists.
type = the file type (PRG, SEQ or USR); optional when reading
or replacing an existing file.
97
mode = type of operation (READ or WRITE); defaults to READ if
not specified.
HINT:
Some or all of the information may be stored in variables. In many situations it is
convenient to store all or part of the file identification information in variables. This
makes it easy for you to write your program so that you can have the user input a file
name. You just store the name in a string variable and then concatenate (add on) the
pieces you need to complete the file identification information. Here are a few
examples of how you can concatenate pieces of information.
OPEN 1,8,5, "SO: MY DATA FILE,S,W" (Original)
INPUT "FILE NAME DESIRED" ;FI* (Name is variable)
OPEN 1,8,5, "@0:"+FI*+",S,W"
(Name, file type and replace
INPUT "NAME OF FILE";NF* are all variable)
INPUT "FILE TYPE (P/S/U)";FT*
INPUT "REPLACE EXISTING FILE (Y/N)";X*
R*="":IF X*="Y" THEN R*="@: "
OPEN 3,8,7,R*+NF*+", "+FT*+",W"
NOTES:
1. To replace an existing file with a new version use @0 as the drive number. Be sure
to read the warnings about replacing files (see Sections 6.6, 7.4 and 8.7).
2. When opening a file to read it, you may use the wild card characters (? and *) in
the name. These are not allowed in write mode.
3. You may use S instead of SEQ, P instead of PRG, and U instead of USR when
specifying the file type.
4. You may use R instead of READ and W instead of WRITE when specifying the
mode of access.
5. There is a special APPEND (A) mode that is used for adding data to the end of an
existing file.
6. There is also a special MODIFY (M) mode that is used for reading unclosed files
(see Section 11.3).
ANNOTATED EXAMPLES:
OPEN 25 , 8 , 4 , " O : BUNCH AD AT A , SEQ , RE AD "
Opens file# 25 to device# 8 (disk) using channel# 4 to access the file on drive zero
called BUNCHADATA. It is a sequential (SEQ) file and we want to read (READ) it.
OPEN 3,8,6, "OrPILAJUNK, SEQ, W"
Opens f ile# 3 to device# 8 (disk) using channel# 6 to access the file on drive zero called
PILAJUNK. It is a new sequential (SEQ) file and we want to write (W) it.
OPEN 1,8,5, "@G: DUMBNAME, P, W"
Opens file# 1 to device# 8 (disk) using channels 5 to replace (@) the file on drive zero
called DUMBNAME. It is a program (P) file and we want to write (W) it.
OPEN 2,8,9, "O: JUNK*, P,R"
Opens f ile# 2 to read (R) a program (P) file whose name begins with the letters JUNK
on drive zero.
OPEN 3,8,6, "O: PILAJUNK"
Opens file# 3 to read an existing sequential file called PILAJUNK. NOTE: this same
command can open an existing program or relative file called PILAJUNK.
Relative Files
The "fid" syntax for relative files is:
SYNTAX:
" dr i ve# : -f i 1 en ame , L , " +CHR* ( 1 en ) (for new or existing files)
"dri ve#: -Fi lename" (for existing files only)
EXAMPLES:
OPEN 1,8,6, "PHONE LIST, L, "+CHR* < 125)
OPEN 7,8,3," XMAS CARDS , L , " +CHR* ( 80 >
OPEN 9,8,12, "CUSTOMERS"
WHERE:
drive# = the drive number (always zero on the 1541); optional on
all Commodore single drives,
filename = the name of the file as given in the directory; may contain
wild cards if file already exists,
len = the record length in characters (2 to 254).
WARNING:
There are three forbidden record lengths. You may not use record lengths of 42, 58, or
63. An OPEN command containing one of these record lengths will be rejected by the
1541 with a SYNTAX ERROR.
CAUTION:
The record length (an integer between two and 254) must be specified when a new
relative file is created. Once you have created the file, its record length cannot be
changed. If you try to specify a new record length for an existing file, your request
will be ignored and the file will not be opened. The only way to change the record
length is to open a new relative file using the new record size and copy the information
into the new file one record at a time!
NOTES:
1. Do not bother trying to replace an existing relative file using @0 as the drive
number. All that happens is a normal open of the file.
99
2. When accessing an existing file, you may use the wild card characters (? and *) in
the name. These are not allowed when creating a new file.
3. You do not have to specify whether you want to read the file or write to a relative
file. Any time you open a relative file you may read or write any record.
ANNOTATED EXAMPLES:
OPEN 5, 8, 4, "O: BIBFILE, L, "+CHR* (45)
Opens file# 5 to device# 8 (disk) using channel* 4 to access an existing relative file
called BIGFILE with a record length of 45, or create a new relative file by this name
if one does not exist.
OPEN 2 1 , 8 , 9 , " O : BL ACKHOLE , L , " +CHR* ( 254 )
Opens file# 21 to device# 8 (disk) using channel# 9 to access an existing relative file
called BLACKHOLE with a record length of 254, or create a new relative file by this
name if one does not exist.
OPEN 6,8,5," BLACKHOLE "
Opens file# 6 to device* 8 (disk) using channel* 5 to access an existing relative file
called BLACKHOLE for read/write access. Note that this can also open a PRG, SEQ
or USR file with this name for read access.
Direct-access Files
The "fid" syntax for direct-access files is:
SYNTAX:
ALTERNATE:
"#buffer"
EXAMPLES:
OPEN 1,8,5, "#"
OPEN 7,8, 12, "#0"
WHERE:
buffer = the buffer number desired (0-2 on the 1541).
NOTE:
The buffer number refers to which part of the disk's RAM you will use. Buffer #0 is
the area $0300-03FF, buffer #1 is $0400-04FF, buffer #2 is $0500-05FF, and buffer
#3 is $0600-06FF. If you do not specify a buffer number, you will be assigned an
inactive buffer by the drive. If no buffer is available, you will get a 70, NO
CHANNEL, error message. For most normal disk operations you don't need to specify
a particular buffer.
ANNOTATED EXAMPLES:
OPEN 5,8,4, "#"
100
Opens file# 5 as a direct-access file using channel# 4. No buffer number is specified.
OPEN 1,8,6, "#2"
Opens file# 1 as a direct-access file using channel* 6. Buffer number two is specified.
Command Channel
The "fid" syntax for the command channel is:
syntax:
nothing
ALTERNATE:
"command"
EXAMPLES:
OPEN 15,8,15
OPEN 15,8, 15, "N: TEST DISK,T5"
WHERE:
command = a disk command such as "10" or "S0:JUNKFILE". See
Appendix D for a complete list of the direct-access
commands. Chapter 4 describes the diskette-
housekeeping commands in detail.
ANNOTATED EXAMPLES:
OPEN 15,8,15
Opens file# 15 as the command channel. No disk command is issued.
OPEN 15,8, 15, "10"
Opens file# 15 as the command channel and sends the disk command "10" to tell the
drive to read the BAM of the diskette in drive zero.
OPEN 3,8, 15, "SO:JUNKYFILE"
Opens file# 3 as the command channel and sends the disk command
"SO:JUNKYFILE" to tell the drive to scratch the file named JUNKYFILE from the
directory of the diskette in drive zero.
5.10 Disk Drive Status
Whenever you perform a disk operation, such as opening a file, the 1541 DOS monitors
the operation. When the operation is completed, the DOS generates a status report
that indicates whether or not the operation was completed successfully. You should
check the disk drive status after any disk operation.
The disk status report consists of four parts: an error code, an English language
message, and the track number and sector number of the block where the error, if
any, occurred. The table below lists the errors most frequently encountered during
normal file handling. A complete listing of all the different error codes and messages
is given in Appendix C.
101
Code
Message
00*
OK
01*
FILES SCRATCHED
02-19*
20
READ ERROR
21
READ ERROR
22
READ ERROR
23
READ ERROR
26
WRITE PROTECT ON
27
READ ERROR
29
DISK ID MISMATCH
30
SYNTAX ERROR
31
SYNTAX ERROR
32
SYNTAX ERROR
33
SYNTAX ERROR
34
SYNTAX ERROR
50*
RECORD NOT PRESENT
51
OVERFLOW IN RECORD
52
FILE TOO LARGE
60
WRITE FILE OPEN
61
FILE NOT OPEN
62
FILE NOT FOUND
63
FILE EXISTS
64
FILE TYPE MISMATCH
66
ILLEGAL TRACK & SECTOR
70
NO CHANNEL
72
DISK FULL
74
DRIVE NOT READY
Meaning
Everything went just fine.
Number of files scratched (not an error).
Undocumented (ignore).
Can't find header of sector.
No sync mark on track requested.
Can't find data block of sector.
Checksum error in the data block.
Attempt write with write protect tab on.
Checksum error in the header block.
Mismatch between BAM & disk ID.
Bad command, general syntax error.
Unrecognized command.
Command longer than 58 characters.
Incorrectly used wild card.
File name omitted from command.
New relative file record being added.
Relative file record too long.
Relative file too large for the disk.
Attempt to read a write file.
File not open.
File with that name does not exist.
File by that name already exists.
File type mismatch.
Pointer to non-existent block.
File did not open.
Disk or directory is full.
No diskette in drive.
* Not errors, therefore may be ignored.
You use the command channel (channel# 15) to check the disk status. It is a good idea
to open the command channel early in your program and leave it open until just
before your program terminates. This allows you to check the disk drive status at any
point in your program.
WARNING:
Do not open and close the command channel each time you want to check the disk
status. Whenever you close the command channel, all files are closed automatically.
This can be used to close a disk file when your program aborts with a SYNTAX
ERROR and you edited the line. See the HINT in Section 3.8 for more details.
To determine the disk drive status, use an INPUT# statement to read the four parts
of the report like this:
10 OPEN 15,8, 15: REM OPEN COMMAND CHANNEL
a • m m
m • * •
230 INPUT#15,E,E*,T,S:REM check disk status
240 IF E>19 THEN ?E;E*;T;S:CL0SE15:ST0P
102
In our example, line 230 inputs the disk status. The first variable, E, will contain the
error code number. The second variable, E$, will contain the error message. The
remaining variables, T and S, will contain the track number (T) and the sector
number (S) where the error, if any, occurred. Line 240 checks the error code number
in E. Error codes less than 20 are ignored. If the error code is greater than 19, the
entire report is printed on the screen, the files closed and execution halted.
5.11 Limitation on the Number of Files Open
If you check the manual for your Commodore 64 or VIC-20, you will find that you are
allowed.to have up to 10 files open simultaneously. For example, you could open one
file to your printer, another one to your modem, another to your cassette tape recorder
and a couple to your disk drive. With all these files open you might get confused, but
your computer wouldn't.
Although your computer can handle 10 open files, the peripheral devices can't. Each
peripheral has its own limitations. Your 1541 is limited by the amount of RAM
memory that it contains. The following table indicates the maximum allowable
number of disk files of various types that you can have open simultaneously.
SEQorPRG Relative Direct Access Total Disk
Files Files Files Files Open
0 0 0-4 0-4
0 1 0-2 1-3
1 1 0-1 2-3
2 0 0-2 2-4
3 0 0-1 3-4
Note that you cannot have two relative files open at once.
5.12 Writing to a File
The PRINT# statement is used to write information into a file. A PRINT# statement
can be used either within a program or in immediate mode. In general it looks like
this:
SYNTAX:
xxx PRINT# -file#, var list
EXAMPLES:
120 PRINT#3,N*3
140 PRINT# 5,M*;
250 PRINT#12,M*+N*", EMPL0YEE#"W1
300 PRINT# 6,NAM*;","; ADR*;","; CITY*
WHERE:
xxx = the line number.
file# = the file number used in the OPEN statement.
103
var list = one or more numeric or string variable names or
expressions; separated, where there may be ambiguity,
by a space, comma or semicolon.
WARNING:
PRINT# does notoutput separators between values. When you use a BASIC command
like X=3:Y=4:PRINT X;Y you see "3 4" on the screen, not "3,4". The PRINT#
command works the same way. This means that when you output the values of
several variables using a single PRINT# statement, there will be no separators
between the values in the file. For example, if A=20, B=5, and C=12 and we use
PRINT#xx,A,B,C, the file will contain 20 5 12 <cr>. This will be hard to read back
correctly. If we use INPUT#xx,X to read this file, the blank spaces will be ignored
and X will be assigned the value 20512. Field separators must be written out explicitly
like this:
PR I NT#x x , A " , " B " , " C (Note the quotes around the commas.)
NOTES:
1. PRINT# is a single BASIC keyword. You may not leave a space between the word
PRINT and the # sign. Spaces between the # sign and the f ile# or between the file#
and the comma (,) are optional.
2. Do notuse ?# as an abbreviation for PRINT#. The correct abbreviation for PRINT#
is pR (press P and then hold the SHIFT key down while pressing R). If you do use
?#, the program will halt with a SYNTAX ERROR but when you list the line, it
will appear correct'
3. PRINTS normally outputs a carriage return character following the value of the
last variable named, unless you specifically suppress it by putting a semicolon at
the end of the variable list.
4. PRINTS spaces the output just like a PRINT statement. The spacing of characters
in the file will be the same as though you used a PRINT statement to print them
on the video screen. This means that each numeric value is preceded by a space
(for the + or - sign) and followed by another space. If you must pack a lot of
numeric values into a file, use the STR$ function to convert them into strings and
then use the MID$ function to delete the leading space like this:
PRINT#x>! , MID* <STR* ( AA) ,2)
This saves two bytes per number.
ANNOTATED EXAMPLES:
120 PRINT#3,A
Writes the value of the variable A followed by a carriage return character into f ile# 3.
470 PRINT#9, A" , "B" , "C*
Writes the value of the variable A, a comma, the value of the variable B, a comma, the
value of the variable C$, and a carriage return character into file# 9.
104
235 PRINT#4, "THIS IS RECORD NUMBER "K
Writes the message THIS IS RECORD NUMBER followed by the value of the
variable K and a carriage return character into file# 4.
125 PRINT#1,MX*<K>?
Writes the value of the Kth element of the one dimensional array named MX$ into
file# 1. No carriage return character is written.
5.13 Monitoring the Number of Blocks Free
Running out of space on a diskette can cause serious problems when you are writing
to a file. You will lose some, if not all, of your data. As a result, it is a good idea to
check the number of blocks free on your diskette whenever you want to write, save,
replace or append new data to a file. This is easy to do if you are in immediate mode.
Just load and list the directory. But how do you do this from within a program? It is
actually quite easy. You can PEEK the value out of the 1541's RAM memory. Type in
this short program and see how it's done.
********************
program: blocks free
********************
0 REM 1541 USER'S GUIDE SECTION 5.13
1 REM COPYRIGHT: G. NEUFELD, 1984
2 :
3 REM READ BLOCKS FREE ON 1541
4 :
10 OPEN 15,8, 15, "10"
20 PR I NT# 15, " M-R " CHR* < 250 > CHR* ( 2 ) CHR* < 3
)
30 GET# 15,A*:BF=ASC<A*+CHR*<0>>
40 GET* 15, A*
50 GET# 1 5 , A* : BF=BF+256* ASC ( A*+CHR* ( O ) )
60 PRINT "THERE ARE"; BF; "BLOCKS FREE"
70 CLOSE 15
Lines 20 to 50 are the ones that PEEK the value you want. You may want to include
these lines as a subroutine in your file handling programs. Whenever you need to
know how much space is left on the diskette, just GOSUB to this routine.
Here are a few suggestions about how to use this routine. When you know the approxi-
mate size of the file that is to be written, you can have your program check to see if
there will be enough space. If there isn't enough space, you might display a message
asking the user to change diskettes. When you don't have any idea about how much
space will be needed, you can have your program monitor the amount of space left as
the file is being written. If you run out of space, you might ask the user to switch
diskettes and then rewrite the entire file. Remember though, when the user switches
diskettes, all files in the 1541 will be closed!
105
5.14 Reading from a File
You can use either an INPUT# or a GET# statement to read information from a file.
In general, the INPUT# statement reads in several characters of information at once
while the GET# statement reads in one character of information at a time. Let's
examine the characteristics of these in some detail.
INPUT# Statement
The INPUT# statement can only be used within a program. It may not be given in
immediate mode. In general it looks like this:
SYNTAX:
xxx INPUT# filett, var list
EXAMPLES:
250 INPUT#1,X
1 00 I NPUT#7 , NUM , NAM* , ADR* , C I T Y* , PROV* , PCODE*
220 INPUT#12,N(K),M*<K+3>
WHERE:
xxx = the line number.
file# - the file number used in the OPEN statement.
var list = one or more variable names separated from each other by
commas. Any combination of numeric or string variables
or arrays may be used.
WARNING:
INPUT# always reads as far as a carriage return character just like an INPUT
statement. Since many people are unaware of the problems this can cause, enter and
run this test program.
********************
PROGRAM: TEST INPUT#
********************
1 0 OPEN 1 f 8 , 5 , " @0 : NUMBERS , S , W " Open file to write.
20 FOR K=l TO 20 Loop 20 times.
30 PRINT#1, K;",";K*K;,,,"K*K*K Write in the number,
40 NEXT its square and cube.
50 CLOSE 1 Close file.
60 OPEN 2, 8, 6," NUMBERS" Reopen to read file.
70 INPUT#2, X Read value from file.
80 PRINT X Display what we read.
90 IF ST=0 GOTO 70 Read to end (Section 5.15).
100 CLOSE 2: END Close file and quit.
The file contains 1, 1, 1 <cr> 2, 4, 8 <cr> 3, 9, 27.... However, when we read it back,
we only get 1, 2, 3, 4,.... The INPUT# statement in line 70 read as far as a carriage
return character each time. It assigned the first value it found to X. Since we did not
106
have any more variable names listed behind the INPUT#, the other two values were
discarded. Notice that we were not warned that values were being discarded. This
can happen whenever you use any delimiter other than a carriage return character as
a field delimiter. Beware!
NOTES:
1. INPUT# is a single BASIC keyword. You may not leave a space between the word
INPUT and the # sign. Spaces between the # sign and the file# or between the
file# and the comma (,) are optional.
2. INPUT# can read past the first carriage return. If your INPUT# statement
specifies several variable names (e.g., INPUT#5,X$,Y,Z) the file will be read until
enough values are found to satisfy all the variables listed (e.g., TIM<cr>15,
29,37<cr>). In this example X$=TIM, Y=15, and Z=39; 37 is ignored (see
WARNING above).
3. Do not read non-numeric characters into a numeric variable. If you specify a
numeric variable name in the INPUT# statement and the computer encounters
any non-numeric characters, the program aborts with a FILE DATA
ERROR message.
4. INPUT# can only handle up to 88 characters. The INPUT# statement on the
VIC-20 and Commodore 64 can handle up to 88 characters between delimiters. If
there are more than 88 characters, the program will abort and report a STRING
TOO LONG ERROR message. If you write programs for other machines, you
should be aware that INPUT# on a 4032 or an 8032 only allows 80 characters
between delimiters.
ANNOTATED EXAMPLES:
120 INPUT#3,A
Reads as far as the next carriage return character in f ile# 3. The first value read is
stored as the value of the numeric variable A.
470 INPUT#9,A,B,C*
Reads at least as far as the next carriage return character in file# 1. The first value
read is stored as the value of the numeric variable A, the second as B, and the third as
C$. If carriage returns are encountered before all three values have been found, the
file will be read as far as the first carriage return character that follows the value
assigned to C$.
125 INPUT#1,MX*<K>
Reads as far as the next carriage return character in file# 1. The first value read is
stored as the value of the Kth element in the one-dimensional array named MX$.
GET# Statement
The GET# statement inputs data one character at a time. It can only be used within a
program. It may not be used in immediate mode. In general, it looks like this:
107
syntax:
xxx GET# file#, var list
EXAMPLES:
120 GET#1,A*
350 GET#5,X*:IF X*="" THEN X*=CHR*(0)
230 GET#12,D*<K)
115 GET#6,A*„B*,C*,D*
WHERE:
xxx = the line number.
file# = the file number used in the OPEN statement.
var list = one or more variable names separated from each other by
commas. Generally only string variables or arrays are
used.
NOTES:
1. GET# is a single BASIC keyword. You may not leave a space between the word
GET and the # sign. Spaces between the # sign and the file# or between the file#
and the comma (,) are optional.
2. Generally use only string variables with GET#. It is possible to use a numeric
variable name in a GET# statement. If you do, the digits zero to nine will be read
correctly, the plus sign (+), minus sign (-), colon (:), comma (,), period (.) and the
letter E will be read as zero, and all other characters will cause a SYNTAX
ERROR.
3. GET# reads a file one character at a time. Carriage return characters, CHR$(13),
commas, colons, and other field or record delimiters are treated the same as any
other character (except CHR$(0) —see below).
3. GET# is slow! Since the computer must send the disk drive a send-me-one-byte
request each time you use GET#, it takes quite a while to read through a file. For
example, when I first wrote the BACKUP program (Appendix E), I used GET# to
fetch the bytes. It took about four hours to backup a diskette! Replacing the GET#
with a short machine language routine cut the time to 25 minutes.
4. CHR$(0) is not read correctly. If your file contains a CHR$(0), it will not be read
correctly. The value stored into the string variable is not CHR$(0). The null string
(no characters) is stored there instead. This is the reason for the common GET#
statement:
6ET#1 , X*: IF X*=" " THEN X*=CHR* <0>
ANNOTATED EXAMPLES:
120 GET#3,A*
Reads one character from file# 3 and assigns its value to the string variable A$.
108
470 GET#9,A*,B*,C*
Reads three characters from file# 9. The value of the first is assigned to the string
variable A$, the second assigned to B$ and the third to C$.
225 GET#5,A*:IF A*="" THEN A*=CHR*(0)
Reads one character from file# 5 and assigns its value to the string variable A$. If a
null byte was read, it is corrected to CHR$(0).
5.15 {Detecting the End of a File
Many people store a special end-of-f ile value, such as -99999, at the end of their data
files so they can tell when they have reached the end of the file. This is not necessary
with your Commodore computer. All you have to do is monitor the value stored in the
special status variable, ST.
The status variable, ST, reflects the status of the most recent input or output (I/O)
operation. Its normal value is zero. As you read the last value in a file, your computer
will set ST to 64. This indicates an end-of-file condition. Type in and run this test
program. The first time choose the W option to create the test file. Once the file has
been stored on diskette, run the program again and choose the R option.
*****************
PROGRAM: ST VALUE
*****************
0 REM 1541 USER'S GUIDE SECTION 5.15
1 REM COPYRIGHT: G. NEUFELD, 19S4
2 :
3 REM OBSERVE THE VALUE OF STATUS VARIAB
LE
4 :
lO INPUT "READ OR WRITE THE FILE (R/W)";
A*
20 OPEN 15,8, 15, "IO"
30 IF A*="R" GOTO HO
40 OPEN 1,8,5, "©O: TEST FILE,S,W"
50 GOSUB 170: IF EN>19 THEN CLOSE 1: CLOSE
15: STOP
60 FOR K=l TO 15
70 PRINT#1,"THIS IS RECORD #";K
80 PRINT "WRITING RECORD #";K
90 NEXT
100 CLOSE l: CLOSE 15: END
HO OPEN 3,8,5, "TEST FILE,S,R"
120 GOSUB 170: IF EN>19 THEN CLOSE 1:CL0S
E 15: STOP
130 INPUT #3, A*
140 PRINT A»;"THE VALUE OF ST ="ST
ISO IF ST=0 GOTO 130
160 CLOSE l: CLOSE 15: END
170 INPUT#15,EN,EM*,ET,ES
109
ISO print en;em*;et;es
190 RETURN
The example below illustrates the use of the status variable to locate the end-of-file
when reading a file.
(File# 1 to disk opened previously.)
200 NUM=0
210 NUM=NUM+1
220 I NPUT# 1 , NAM* ( NUM ) , ADR* ( NUN )
230 PRINT NAM* < NUM), ADR* <NUM)
240 IF ST<>64 GOTO 210
In our example above, we read data from the file, printed it to the screen and then
checked the status variable, ST, to see if we were at the end of the file. Note that
printing the data to the screen did not affect the value of ST. However, if we had
printed the data to a printer using a PRINT# statement in line 230, we would be in
trouble. The value of ST in line 240 would then indicate the status of the printing
operation in 230 rather than the input operation in 220. When we reached the end of
the file, we would never know it. In situations such as this, we have to save the value of
ST before it is lost like this:
(File# 1 to disk and file# 4 to printer opened previously.)
200 NUM=0
210 NUM=NUM+1
220 I NPUT# 1 , NAM* ( NUM ) , ADR* ( NUM )
225 S=ST:REM SAVE STATUS
230 PR I NT#4 , NAM* ( NUM ) , ADR* ( NUM )
240 IF S<>64 GOTO 210
CAUTION:
Even using INPUT# to read the disk drive error status (see Section 5.10) will change
the value of ST.
5.16 Closing a File
The CLOSE statement is used to close a file when you are finished using it. It is very
important to close a file, particularly if you have been writing to it. Otherwise you
may create an unclosed file.
A CLOSE statement looks like this:
SYNTAX :
xxx CLOSE file#
EXAMPLES:
50 CLOSE 1
200 CLOSE 5: CLOSE 15
600 CLOSE CC
110
WHERE:
xxx = the line number.
file# = the file number used in the OPEN statement.
WARNINGS:
1. You will lose data if you don't close a write file! If you fail to close a sequential
(SEQ), program (PRG) or user (USR) file that you have just written, you must use
the techniques described in Section 11.3 to be able to access any of your data at all.
Even using these techniques, you will lose data! If you fail to close a relative file
after writing to it, some of the data will probably be lost. Be safe, not sorry! Close
those files.
2. Closing the command channel will close all open files. Whenever you close the
command channel, all the disk files that are open are closed automatically. This is
the reason for the rule that the command channel should be opened first and
closed last. See the HINT in Section 3.8 for a useful application of this.
ANNOTATED EXAMPLES:
120 CLOSE 3
Closes down file number three.
100 FL - 12
■ • ■
m m m
470 CLOSE FL
Closes file number 12.
5.17 Conclusion
Now that you have learned some of the basic concepts of file handling and the
commands you will need, it's time to get down to specifics. The next three chapters
take a detailed look at program, sequential and relative files. Each chapter contains
specific suggestions and examples to help you make more effective use of one type of
file. Each of these chapters stands alone. You do not have to read and understand
Chapter 6 to make use of Chapter 7 or 8.
Ill
CHAPTER
6
PROGRAM FILES
This chapter takes a detailed look at program files and their contents. It includes an
introduction to the inner structure of BASIC programs and concludes with a look at
self-modifying programs. Each section includes several examples to help you learn
how to make better use of this common, but underutilized, file type.
6.1 Introduction to Program Files
A program file is the most common type you will encounter. It is used to store a copy
of the contents of some part of the computer's RAM or ROM memory. Usually it holds
a copy of a BASIC or machine language program. Your Commodore 64 makes it very
easy for you to create or read program files that contain BASIC programs. All you do
is use a SAVE or LOAD command. When you do this the program is saved or loaded
faster than any other file type.
Normally program files are not used for data storage. However some application
programs, such as the word processing programs WordPro™ and PaperClip™, use a
program file to store data. In these cases, the data (the text) always resides in a
particular part of the computer's memory. When you want to memorize (save) the
text you have typed in, the program sets some pointers and uses a SAVE command to
save a copy of the appropriate part of the computer's RAM memory to the diskette.
When you want to recall the text, the program uses a LOAD command to load the
program file from diskette into the computer's memory- These programs use program
files for data storage because loading or saving a program file is faster and easier
than reading or writing a data file.
You are not limited to just saving or loading a program file. You can open the file and
read or write to it as well. This is done to create a compacted program (all REM
statements and unnecessary spaces are deleted), a cross reference (a list of the line
numbers in which each variable is used), or a specialized program listing (such as
was used to create the listings for this book).
The biggest difficulty in using program files in nonstandard ways is that you must
have a thorough understanding of the way in which a BASIC program is structured
and tokenized in your computer's memory. You must be sure to get all the pointers
and tokens correct or your computer will hang.
113
When to use a program file:
1. To store BASIC or machine language programs in a form that can be loaded
easily.
2. To store blocks of data that always occupy the same area of the Commodore 64's
RAM memory (advanced users only).
The benefits from using program files are:
1. They are easy to access using SAVE or LOAD.
2. They are saved or loaded at maximum speed.
The drawbacks of using program files are:
1. The amount of information you can store in a program file is usually limited to the
amount that can be held in the computer's memory at one time.
2. You must understand how a BASIC program is stored in memory to make sense
of the contents of a program file or to create a new program file.
Some typical uses of program files are:
1. Storing BASIC programs.
2. Storing machine language programs.
3. Storing text by word processors.
6.2 Structure of a Program File
A program file has a simple structure. It is not divided into fields or records. The first
two bytes in the file specify where the file is to start loading into memory (the load
address). The rest of the file is a copy of what is to be loaded into memory. A program
file that contains a BASIC program for the Commodore 64 would look like this:
$01 $08 < BASIC program
The load address consists of two bytes (in this file $01 and $08). The least significant
byte is stored first (lo byte/hi byte). In this case the load address is $0801. You can
convert this into decimal notation as follows:
load address = 256 * hi byte + lo byte
load address = 256 * 8 + 1 = 2049 ($0801)
A typical machine language routine looks like this:
$00 $25 < Machine language program
load address = 256 * (2*16 + 5) + 0 = 9472 ($2500)
114
6.3 Loading a Program File
You may load a copy of a program file into the Commodore 64's memory from
immediate mode (i.e., you have a READY prompt and a flashing cursor),or from
within a program.
Loading from Immediate Mode
Type in one of the following commands and press RETURN.
SYNTAX:
LOAD "filename", device* (program relocated)
LOAD "filename", device#, channel* (not relocated)
EXAMPLES:
LOAD "GOOD GAME", 8
LOAD "ML SORT ",8,1
F*=" SUPER FORCE"
LOAD F*,8
WHERE:
filename = the name of the program file as it appears in a directory
listing.
device# = the device number of the 1541 (normally eight),
channel* = the channel number (secondary address). (Normally one
for a non-relocating load; used when loading machine
language programs.)
NOTES:
1. The file name may be stored as a string variable. If it is, use the string variable
name instead of "filename."
2. The file name may contain one or more wild cards (? or *). If it does, the first file
name that matches the pattern will be loaded into memory.
3. If a program file with the name given is not stored on the diskette, you will get a
63, FILE NOT FOUND, error message.
REMINDER:
If you do not specify ",1" after the device number (eight) in a LOAD command, the
Commodore 64 will ignore the load address at the start of the file. The program Will
be loaded into memory starting at memory location 2049 (where a BASIC program
normally starts in the Commodore 64).
ANNOTATED EXAMPLES:
LOAD "SPACE JUNK", 8
Loads a copy of the program file, SPACE JUNK, into RAM memory starting at
location 2049.
115
LOAD"*", 8
Loads a copy of the directory file. The disk drive converts this into a pseudo program
so you can list it. Do not use a ",1" when loading the directory. It has a load address of
1025 ($0401). If you use the ",1" form, the directory will be loaded into your
Commodore 64's screen RAM and produce garbage on the screen.
PN*=" UPPER CODE"
LOAD PN* ,8, 1
Loads a copy of the program file, UPPER CODE, into RAM memory starting at the
load address specified at the start of the program file.
Loading from Within a Program
You can include a LOAD command within a BASIC program. When you do this, you
do not have to RUN the incoming program. It is automatically run from the beginning
as soon as the load is complete. This technique is often used:
1. To load one of several BASIC programs stored on a diskette in response to the
user's selection from a menu.
2. To link or chain several program segments together when your program is too
long to be held in memory at once.
3. To load one or more machine language routines (such as a fast sorting routine)
needed by the BASIC program.
To have your program load another program or a machine language routine, simply
include a line like this:
SYNTAX:
xx LOAD "-f i 1 ename" , devi ce# (program relocated)
xx LOAD "f i 1 ename", device* , channel # (not relocated)
EXAMPLES:
1 50 LOAD " GOOD GAME " , 8
300 LOAD "ML SORT ",8,1
120 F*=" SUPER FORCE"
130 LOAD F*,8
WHERE:
xx = the line number.
filename = the name of the program file as it appears in a directory
listing.
device# = the device number of the 1541 (normally eight).
channel# = the channel number (secondary address; normally one for a
non-relocating load).
116
WARNING:
The first program must be the longest. Any BASIC program that you load from
within another program must not be longer than the original program. The reason
for this is that the variables are stored immediately following the program in RAM
memory. If you were permitted to load in a longer program, the values of the variables
would be destroyed. This can pose a problem if you want to create a short menu
program that loads one of several long programs, unless you trick the Commodore 64
into believing that the menu program is very long by changing the end-of-program
pointer like this:
10 REM MENU PROGRAM
20 P0KE45,XX:REM SET LO BYTE OF POINTER
30 P0KE46,YY:REM SET HI BYTE OF POINTER
40 CLR:REM RESET THE OTHER POINTERS
50 REM PROGRAM CONTINUES...
Get the XX and YY values by PEEKing memory locations 45 and 46 when the
longest program is in memory.
NOTES:
1. The notes that apply to loading from immediate mode also apply to loading from
program mode.
2. The values of variables are preserved. The values of any variables used in your
program are not erased when you load in another BASIC program or machine
language routine! The only time that values may be lost is when you have assigned
a value to a string variable in your BASIC program like this: 5 A$- 'VALUE". In
this situation, the value of A$ is not moved up into high RAM. The pointer to the
value of A$ points at the word "VALUE" within the program. If you load in
another program, the value of A$ will be lost. To prevent this, you must perform
some string manipulation with A$ such as 5 A$="VALUE"+" ". This forces the
value of A$ to be stored in high memory. Now your starting program will pass all
its variables' values to subsequent programs.
3. After a LOAD, your program runs from the beginning. Whenever you load a
program file, the BASIC program in memory will immediately begin to RUN
from the beginning. This is exactly what you want if you just loaded a new BASIC
program. However, if you load a character set or machine language routine into
high memory, your BASIC loader program will be run again from the beginning.
If you do not take steps to handle this, you'll end up in a loop that loads and reloads
your routine endlessly.
(produces endless loop) (prevents a loop)
10 LOAD"MACHINE CODE", 8,1 10 IF X=l GOTO 30
20 X-l: LOAD "MACHINE CODE", 8,1
30 REM PROGRAM CONTINUES. . .
117
4. Loading machine code can cause your computer to hang. Whenever you load a
program file, the end-of-program pointer (memory locations 45 & 46) in your
Commodore 64 is reset to point to the end of the program or routine that you just
loaded. When you load a routine into high memory, the end-of-program pointer
will end up pointing to the end of memory. The first time that you try to store any
values in memory, you will get an ?OUT OF MEMORY ERROR. The trick is to
preserve the end-of-memory pointers before you LOAD and restore them im-
mediately afterwards. Memory locations 1020-1023 are not used by BASIC and
are very handy for temporary storage like this:
lO IF X=l GOTO 50
20 POKE 1020, PEEK (45): REM SAVE LO BYTE
30 POKE 1 021, PEEK ( 46 ): REM SAVE HI BYTE
40 X=l:LOAD"MACHINE CODE", 8,1
50 POKE 45, PEEK (1020): REM FIX LO BYTE
60 POKE 46, PEEK (1021): REM FIX HI BYTE
ANNOTATED EXAMPLES:
15 IF A=l THEN LOAD "SPACE JUNK", 8
Loads a copy of the program file, SPACE JUNK, into RAM memory starting at
location 2049 and runs it.
70 INPUT "FILE NAME";F*
80 LOAD F*,8, 1
Loads a copy of the program whose name was input in line 70. The file is loaded into
RAM memory starting at the load address specified at the start of the file. When the
load is complete, the BASIC program in memory is run from the beginning.
6.4 Saving a Program File
You may save a copy of the contents of part of your computer's memory from
immediate mode (i.e., if you have a READY prompt and a flashing cursor), or from
within a program.
Saving from Immediate Mode
Type in the following command and press RETURN.
SYNTAX:
SAVE "-Filename", device*
EXAMPLES:
SAVE "NEW GAME", 8
F*=" SUPER SOUND"
SAVE F*,8
WHERE:
filename = the name of the program file as you want it to appear in a
directory listing.
device# = the device number of the 1541 (normally eight).
118
WARNING:
The file name must not contain any of the following characters: asterisk (*), colon (:),
question mark (?), quotation mark ("), or comma(,). Graphics characters, reverse-field
characters and other special characters are permitted but their use is discouraged.
NOTES:
1. The file name may be stored in a string variable. If it is, use the string variable
name instead of "filename."
2. The file name may be up to 16 characters long.
3. If a program file with the name given is already stored on the diskette, you will
get a 62, FILE EXISTS, error message.
HINTS:
1. Choosing a name. You may save your program using any name you like. The disk
drive does not care whether the name you select has any relationship to the nature
of the program. But, to save you endless hours of frustration, choose a name you
can remember. Otherwise as your collection of diskettes grows, you will be unable
to find anything.
2. Saving machine language routines. The SAVE command stores a copy of RAM
memory onto a diskette. The first byte that will be stored is one pointed to by the
start-of-BASIC pointer (memory locations 43 and 44). The start-of-variables pointer
(45 and 46) points to the last byte to be saved plus one. If you reset the pointers
before you save, you can save a copy of any area of RAM. For example, to save a
copy of the video screen RAM onto diskette type in:
P0KE43 , O : P0KE44 , 4 : P0KE45 , 232 : P0KE46 , 7
SAVE" SCREEN " , 8
WARNING:
When the save is complete, either POKE these pointers with their original values or
turn OFF your Commodore 64. . , ;£
REMINDER:
Watch the error light on your drive. Occasionally a program will not be saved properly
onto the diskette. When this happens, the red light on the drive will flash rapidly. Do
not panic! Just try again using a different diskette. The most common causes for this
are the diskette already has a file with the name that you chose, the diskette is full or
the diskette is flawed. After your program has been saved correctly, be sure to check
the directory of the diskette on which you had problems. If there is an unclosed file in
the directory (*PRG), don't scratch it; VALIDATE the diskette (see Section 4.8).
ANNOTATED EXAMPLES:
SAVE "SPACE JUNK", 8
Saves a copy of the program in memory onto diskette using the name SPACE JUNK.
119
A*= " BUNCHASTUFF " : SAVE A* , 8
Saves a copy of the program in memory onto diskette using the name
BUNCHASTUFF.
Saving from Within a Program
You can include a SAVE command within a BASIC program. This is not generally
very useful because all you get on disk is a copy of the existing program. The only
times this is useful are when you have a program that modifies itself as it runs (the
MAIL LIST in Section 6.11 does this) or when you change the pointers and save a
copy of something other than the program itself.
To have your program save a copy of itself at any point in the program, just include a
line like this:
SYNTAX:
xx SAVE "filename", 8
examples:
125 SAVE "NEW GAME", 8
300 F*=" SUPER SOUND"
310 SAVE F*,8
WHERE:
xx = the line number.
filename = the name of the program file as you want it to appear in a
directory listing.
NOTE:
The same warnings, notes, and reminders that apply to saving from immediate mode
also apply here.
HINT:
Saving machine language routines. Many people believe you must be in immediate
mode and use a machine language monitor such as MICROMON or SUPERMON to
save anything other than a BASIC program. This is not truelYou can manipulate the
start-of-B ASIC and start-of-variables pointers so that you can save any part of memory
you want. The part of memory to be saved might contain such things as a character
set, a sprite character or a machine language routine. You can do this either in
immediate mode or from within a BASIC program. To pull off this bit of magic in a
program you have to be careful. Here's what to do:
1. Save the current values of the pointers to the start of BASIC (memory locations 43
& 44) and to the start of the variables (memory locations 45 & 46) by POKEing the
values into some safe area of RAM (1020-1023 is just fine).
2. Set the start-of-BASIC pointer to point to the first byte you want saved (start-of-
BASIC pointer = first byte saved).
120
3. Set the start-of-variables pointer to point to the last byte you want saved plus one
(start-of-variables pointer = last byte saved + 1).
4. Issue a normal SAVE command.
5. Reset the pointers. WARNING: Do not use any variables for any purpose until the
pointers have been reset!
ANNOTATED EXAMPLES:
15 IF A=l THEN SAVE "SHOOT UP", 8
Saves a copy of the program in memory onto diskette using the name SHOOT UP.
********************
PROGRAM: SCREEN SAVE
********************
0 REM 1541 USER'S GUIDE SECTION 6.4
1 REM COPYRIGHT: G. NEUFELD, 1984
2 :
3 REM SAVE A COPY OF THE C-64'S SCREEN R
AM
4 :
10 REM SAVES A SCREEN IMAGE
20 PRINT" «2LR>": REM CLEAR THE SCREEN
30 F0RK=1024T02023: POKEK, INT (256*RND < 1 ) )
: NEXT
40 POKE 1020, PEEK (43): POKE 1021, PEEK (44)
:rem save start
50 POKE 1022, PEEK (45): POKE 1023,PEEK(46)
:REM SAVE END
60 POKE 43,0: POKE 44. 4: REM SET START TO
START OF SCREEN
70 POKE 45, 233: POKE 46, 7: REM SET END TO
END OF SCREEN
80 SAVE "SCREEN IMAGE", 8
90 POKE 43, PEEK (1020): POKE 44, PEEK ( 1021)
:REM RESET START
lOO POKE 45, PEEK (1022): POKE 45, PEEK (1023
):REM RESET END
6.5 Verifying a Program File
After you have saved or loaded a copy of a program you may check to see if the copy in
RAM memory matches the copy on diskette by using the VERIFY command. Usually
a program is verified from immediate mode (i.e., you have a READY prompt and a
flashing cursor). You can do a verify from within a program but this has virtually no
practical value.
121
Verifying from Immediate Mode
Type in one of the following commands and press RETURN.
SYNTAX:
VERIFY "-Filename",device# (program relocated)
VERIFY "-f ilename", device*, channel # (not relocated)
If there is an exact match between the copy on diskette and the copy in RAM, the
message OK will be displayed. If there is not an exact match, the message ? VERIFY
ERROR will be displayed.
EXAMPLES:
VERIFY "ADVENTURE", 8
VERIFY "SUPER SORT ",8,1
F*=" INCOME TAX"
VERIFY F*,8
WHERE:
filename = the name of the program file as it appears in a directory
listing. Wild cards may be used,
device* = the device number of the 1541 (normally eight).
channel# = the channel number (secondary address; normally one for
a non-relocating verify).
WARNING:
If you do not specify ",1" after the device number (eight) in a VERIFY command, the
Commodore 64 will ignore the load address at the start of the file. The contents of the
diskette file will be compared to the contents of memory starting at memory location
2049, where programs normally start on the Commodore 64.
NOTES:
1. The file name may be stored in a string variable. If it is, use the string variable
name instead of "filename."
2. The file name may contain one or more wild cards (? or *). If it does, the first file
name that matches the pattern will be used for verifying.
3. If a program file with the name given is not stored on the diskette, you will get a
63, FILE NOT FOUND, error message.
REMINDER:
If you have used the LOAD "filename",8 form of the LOAD command to load a
BASIC program that was originally saved from a PET or a VIC-20, you will always
get a VERIFY ERROR when you try to verify it. The reason is that during the load,
the line links (see Section 6.7) have been changed so that the program will work
correctly on your Commodore 64. There is no way to verify this kind of file!
122
ANNOTATED EXAMPLES:
VERIFY"SPACE JUNK", 8
Compares the copy of the program, SPACE JUNK, on diskette with the program
that starts at 2049 in RAM memory.
FI*="FANCY STUFF"
VERIFY FI«,8, 1
Compares the copy of the program, FANCY STUFF, on diskette with the area of
RAM memory starting at the load address specified at the start of the file.
Verifying from Within a Program
You can include a VERIFY command within a BASIC program. The only real use
for this would be to check if a machine language routine or character set was loaded
or saved properly. The syntax is the same as from immediate mode except for the line
number. If the copies match, an OK message is printed on the screen and execution
continues. If the copies don't match, the program aborts with a VERIFY ERROR
message. If a copy of the named file is not on the diskette, the program aborts with a
FILE NOT FOUND error.
6.6 Replacing a Program File
You may replace an existing copy of a program file on disk with a revised copy of the
Commodore 64's memory from immediate mode (i.e., you have a READY prompt
and a flashing cursor) or from within a program.
Replacing from Immediate Mode
Type in the following command and press RETURN.
SYNTAX:
SAVE "@: -filename", device*
ALTERNATE:
SAVE "@0:-filename",device#
EXAMPLES:
SAVE "@:GOOD GAME", 8
F*="@0:NEW STUFF"
SAVE F*,8
WHERE:
filename = the name of the program file as it appears in a directory
listing.
device# = the device number of the 1541 (normally eight).
WARNING:
Be sure there is enough disk space to do a replace. When the file is opened for a
replace, the old file is not deleted before the new file is written. This means that you
must have enough free space on the disk for the new copy. If there is not enough
123
space, the replace is aborted. The new file is incomplete and the old file is unrecov-
erable, unless you use the techniques given in Section 11.8. Be safe, not sorry! Check
the number of blocks free on your disk before you try a replace (see Section 5.13).
NOTES:
1. The file name may be stored in a string variable. If it is, use the string variable
name instead of "filename."
2. The file name may contain one or more wild cards (? or *). If it does, the first file
name that matches the pattern will be replaced by the current memory contents.
3. If a program file with the name given is not stored on the diskette, you will get a
63, FILE NOT FOUND, error message.
REMINDER:
Do not replace an unclosed file! Be sure that the original program file you are going to
replace has been properly closed. A file that has not been properly closed has an
asterisk beside its file type in the directory (*PRG). Use the VALIDATE command
(see Section 4.8) to eliminate these files! If you accidentally replace one of these files,
VALIDATE the diskette as soon as possible.
ANNOTATED EXAMPLE:
S AVE " @0 : SPACE J UNK " , 8
Replaces the existing copy of the program, SPACE JUNK, with a copy of the current
program in RAM memory.
Replacing from Within a Program
You can include a SAVE with replace command within a BASIC program. This
works the same as a SAVE within a program. Use the techniques outlined in Section
6.4 but put an @: or @0: in front of the file name.
NOTES:
1. Unlike the ordinary SAVE command, the file name in the replace command may
use one or more wild cards (? or *). If it does, the first file name that matches the
pattern will be replaced by the current memory contents.
2. Program execution will continue with the next line.
6.7 The Internal Structure of a BASIC Program
To be able to use a program file for more than simply saving or loading programs you
need to know how a BASIC program is put together. You may be surprised to
discover that a program is not stored in memory the same way you typed it in or the
way you see it when you LIST it. In general, one line of a BASIC program in memory
looks like this:
124
Pointer to
next line
Line
Number
BASIC statement with
keywords token i zed
Null \Next line
byte ? follows
The first line of a BASIC program in your Commodore 64 begins at memory location
2049 ($0801). Its length is determined by what you typed in. It ends with a null byte
($00). The second line immediately follows the first. All the lines in a BASIC program
are stored in this same manner.
HINT:
At this point you may want to let your computer help you. If you type in the program,
PRG HEX DUMP (see Appendix E), it will help you understand how a program is
organized. To use the program, save the program you want to analyze on diskette.
Load and run PRG HEX DUMP. When you are asked for the name of the program
you want analyzed, just type in the name you used to save it. PRG HEX DUMP will
access the program file on diskette and produce a hex dump of the program with the
important features highlighted. PRG HEX DUMP is fairly long but typing it in will
be worth the time and effort it takes. You will have a utility that will allow you to scan
any program. It flags any bad line numbers, line links or illegal bytes that it finds If
you don't want to type in PRG HEX DUMP, you can use a machine language monitor
such as SUPERMON, TINYMON or MICROMON to follow along.
To see how this works in practice let's examine how a simple four line program is
stored in the Commodore 64's memory. Here is our demonstration program.
10 REM SAMPLE PROGRAM
20 FOR K=l TO 20
30 PRINT K;K*K
40 NEXT
Here is a dump of the Commodore 64's memory when the sample program shown
above has been typed in.
. : 0800 00 16 08 OA 00 8F 20 53 10 REM SAMPLE PROGRAM
.: 0808 41 4D 50 4C 45 20 50 52
. : 0810 4F 47 52 41 4D 00 25 08
.: 0818 14 OO 81 20 4B B2 31 20 20 FOR K=l TO 20
.: 0820 A4 20 32 30 00 31 08 IE
.: 0828 00 99 20 4B 3B 4B AC 4B 30 PRINT K;K*K
. : 0830 00 37 08 28 00 82 00 00 40 NEXT
. : 0838 00 xx xx xx xx xx xx xx
Let's look at how the start of this program is organized.
1. The null byte just before the start of a BASIC program:
. : 0800 H 16 08 OA 00 8F 20 53 10 REM SAMPLE PROGRAM
. : 0808 41 4D 50 4C 45 20 50 52
. : 0810 4F 47 52 41 4D 00 25 08
125
The first byte shown in the hex dump of the Commodore 64's RAM memory, $0800
(2048), is not part of the BASIC program. However, it is important to the proper
functioning of our program. Unless this memory location is $00, the program will list
but will not ruril Note that this byte is not saved when you save the program.
2. The pointer to the next line:
. : 0800 OO 16 OB OA 00 8F 20 53 10 REM SAMPLE PROGRAM
.: 0808 41 4D 50 4C 45 20 50 52
. : 0810 4F 47 52 41 4D 00 111 08
points to ¥
The first two bytes in a line are a pointer that points to the first byte in the next
program line. The pointer is in lo byte/hi byte form. In this case the pointer points to
$0816. If you are more comfortable with decimal notation, you can find the memory
location that it points to using this formula.
pointer value= 256 * hi byte + lo byte
pointer value= 256 * 08 + 22 = 2118
3. The line number:
. : 0800 00 16 08 WBSM 8F 20 53 10 REM SAMPLE PROGRAM
. : 0808 41 4D 50 4C 45 20 50 52
.: 0810 4F 47 52 41 4D 00 25 08
Bytes three and four in a line are the line number. They are stored in lo byte/hi byte
form. In this case the line number is $000A. If you are more comfortable with
decimal notation, you can find the line number using this formula.
line number = 256 * hi byte + lo byte
line number = 256 * 00 + 10 = 10
4. The BASIC statement with keywords tokenized.
.: 0800 00 16 08 OA 00 8F 20 53 10 REM SAMPLE ROGRAM
.: 0808 41 40 £0 4C 45 20 5C 52
.: 0810 4F 47 52 41 4D 00 !?S5 OS
NOTE:
$8F = BASIC token for REM (more about tokens below).
To save space in memory the BASIC keywords such as REM, PRINT, etc. are
represented by a single character called a BASIC token. BASIC tokens have numeric
values between $80 and $DA (128 to 218). Numbers, letters and spaces within a line
126
are represented by their ASCII codes. Graphics characters are not legal BASIC
characters and may only appear between quotes. A list of all the BASIC tokens and
ASCII codes is given in Appendix B.
5. The end-of-line null byte:
.: 0800 00 16 08 OA OO 8F 20 53
.: 0808 41 4D 50 AC 45 20 50 52
.: 0810 4F 47 52 41 4D 00 25 08
The last byte in a line of a BASIC program is always a null byte ($00).
Now that we've seen how the program is organized, let's take apart the program
character by character to see how it is stored in the Commodore 64's memory.
HINT:
dd^atPt0*1"* y°U may Want to let your comPuter helP you- If you type in the program
PRG ANALYZER, listed in Appendix E, you can use it to get a character by character
analysis of any program. To use this program, save the program you want to analyze
on disk. Then load and run PRG ANALYZER. It will ask you for the name of the
program you want analyzed. It will then give you a complete character by character
analysis of your program. This is quite a long program, but it is a very useful utility.
Here's the hex dump of our sample program again.
.: 0800 00 16 08 OA 00 8F 20 53 10 REM SAMPLE PR08RAM
. : 0808 41 4D 50 AC 45 20 50 52
.: 0810 4F 47 52 41 4D 00 25 08
.: 0818 14 00 81 20 4B B2 31 20 20 FOR K=l TO 20
-: 0820 A4 20 32 30 00 31 08 IE
.: 0828 00 99 20 4B 3B 4B AC 4B 30 PRINT K;K*K
.: 0830 00 37 08 28 00 82 00 00 40 NEXT
. : 0838 00 xx xx xx xx xx xx xx
Remember that the first byte of our dump, $00, at $0800 (2048), is not actually part of
the program. However, if this memory location is not $00, the program will LIST but
will not RUN. If you try to run it, all you will get is a ?SYNTAX ERROR message.
The first line of our program is 10 REM SAMPLE PROGRAM. Our PRG
ANALYZER indicates that it is stored from $0801 to $0815 like this:
$0801/2 $16 $08 pointer to $0816, the start of line 20
$0803/4
$0A $00
line number (10)
$0805
$8F
BASIC token for REM
$0806
$20
ASCII code for space
$0807
$53
ASCII code for S
$0808
$41
ASCII code for A
$0809
$4D
ASCII code for M
$080A
$50
ASCII code for P
$080B
$4C
ASCII code for L
127
$080C
$45
$080D
$20
$080E
$50
$080F
$52
$0810
$4F
$0811
$47
$0812
$52
$0813
$41
$0814
$4D
$0815
$00
ASCII code for E
ASCII code for space
ASCII code for P
ASCII code for R
ASCII code for O
ASCII code for G
ASCII code for R
ASCII code for A
ASCII code for M
end of line marker
The second line of our program is 20 FOR K=l TO 20. It is stored from $0816 to $0824
like this:
pointer to $0825, the start of line 30
line number (1*16+4=20)
BASIC token for FOR
ASCII code for space
ASCII code for K
BASIC token for =
ASCII code for 1
ASCII code for space
BASIC token for TO
ASCII code for space
ASCII code for 2
ASCII code for 0
end of line marker
The third line of our program is 30 PRINT K;K*K. It is stored from $0825 to $0830
like this:
pointer to $0831, the start of line 40
line number (1*16+14=30)
BASIC token for PRINT
ASCII code for space
ASCII code for K
ASCII code for ;
ASCII code for K
BASIC token for *
ASCII code for K
end of line marker
pointer to $0837, the end of the program
line number (2*16+8=40)
BASIC token for NEXT
end of line marker
128
$0816/7
$25 $08
$0818/9
$14 $00
$081A
$81
$081B
$20
$081C
$4B
$081D
$B2
$081E
$31
$081F
$20
$0820
$A4
$0821
$20
$0822
$32
$0823
$30
$0824
$00
$0825/6
$31 $08
$0827/8
$1E $00
$0829
$99
$082A
$20
$082B
$4B
$082C
$3B
$082D
$4B
$082E
$AC
$082F
$4B
$0830
$00
The fourth line of our pr<
$0831/2
$37 $08
$0833/4
$28 $00
$0835
$82
$0836
$00
The last two bytes in our program are the two $00 bytes at $0837 and $0838. These
two zeros are in the space normally occupied by the pointer to the start of the next
line. This signals the end of your program.
Now that you understand how a BASIC program is put together, you will be able to
create programs that read or even create other programs.
HINT:
Have you ever loaded a program, listed it, and found to your dismay that the line
numbers or the line pointers have become jumbled? This can happen because of a bad
load or POKEs to the wrong memory locations. Many people give up in disgust when
this happens. Now that you understand how a BASIC program is put together, you
can fix it using a monitor program. However, you still have to locate the problem.
Here's how to do it.
Load and run your machine language monitor program. Exit from the monitor and
load in the problem program. Now type in the following lines (no line numbers
please).
S=2049
F0RK=1T01000: M=PEEK (S) +256*PEEK ( S+ 1 ) : L=P
EEK (S+2) +256*PEEK <S+3) : ?L; M: S=M: NEXT
You will get two lists of numbers like this.
10 2053
20 2096
30 2120
The number on the left is the line number. The number on the right is the pointer to
the start of the next line. When the numbers in either column are out of sequence, you
have found your problem. The last number that is in the correct sequence in the
second column will point to the problem area. Now that you have located the problem
use your monitor and knowledge of how the BASIC pointers work to solve the problem.
6.8 Using a Program File
To be able to write programs that manipulate existing program files or even create
new ones, you need to know how to read or write a program file. It is really quite easy.
Follow these steps:
1. Open the program file.
2. Write information into the file using PRINT#.
3. Read information from the file using GET#.
4. Close the program file when you are finished using it.
Let's look at each of these steps in more detail.
129
Opening a Program Hie
An OPEN command is used to tell the disk drive the name of the file you wish to use.
On the Commodore 64 or the VIC-20, the OPEN command to access a program file
looks like this:
syntax:
OPEN file#, device**, channel #, "-f ilename,PRG,mode"
EXAMPLES:
OPEN 5,8,4,"C0PY/ALL,PRG,R"
OPEN 1,8,5," BUNCH ADAT A , P , R "
F*=" SUPER FORCE"
OPEN 2,8, 12,F*+",P,W"
WHERE:
file# = the logical file number (any integer 1-127).
device# = the device number (normally eight on the 1541).
channel# = the channel number (any integer 2-14).
filename = the file name (up to 16 characters), may be contained in a
string variable.
PRG = file type, may be abbreviated with P.
mode = R or W to indicate whether you want to read from a file or
write to a file.
NOTES:
1. Part or all of the file identification information may be stored in a string variable.
2. The file type and mode are optional. If the file type is omitted, the file type will be
determined from the directory entry for the file. If the mode is omitted, it defaults
to READ mode unless the file is a relative file. In this case READ/WRITE mode
is assumed.
3. You can open a program file in APPEND (A) mode to add new information to the
end of the program file. The only possible use for this mode appears to be to
append a machine language routine onto the end of a BASIC program. For more
information about APPEND mode see Section 7.5.
4. There is a special MODIFY (M) mode for reading an unclosed file. See Section
11.3 for more details.
READ MODE:
5. The file name may contain one or more wild cards (? or *). If it does, the first file
name that matches the pattern will be opened.
130
6. If there is no file with the name specified stored on the diskette, you will get a 63,
FILE NOT FOUND, error message.
WRITE MODE:
7. The file name may not contain any wild cards (? or *).
8. If there already is a file with the specified name stored on the diskette, you will get
a 62, FILE EXISTS, error message. To write over (replace) an existing file see
below.
REMINDERS:
1. Monitor the error status of the disk drive. You should use the command channel to
monitor the error status of the disk drive. Be sure to check the status whenever
you open a file to make sure that the file has been opened properly. If the file is not
open, your new data will be lost! See Section 5.10 for this procedure.
2. OPEN the command channel first and close it last Do not open and close the
command channel each time you want to check the error status. Open the command
channel at the beginning of your program and leave it open until the end. When
you CLOSE the command channel all open files are closed automatically.
3. All open files must have different logical file numbers. Since the logical file
number is used to identify one particular file, you must use different logical file
numbers for each of the files that are open simultaneously. Once you close a file,
you are free to reuse that logical file number in your program.
4. All open files must have different channel numbers. Since the channel number
(secondary address) identifies a particular data communication channel in the
disk drive, each file that is open must have a unique channel number. Once you
close a file, you are free to reuse that channel number in your program.
5. You can only have a few files open at once. Each sequential or program file that
you have open takes up one of the buffers in the disk drive. Since the disk has only
four buffers and one is reserved for internal use, you can only have three files open
at once (not five as indicated on page 45 of your 1541 manual)! See Section 5.11 for
the limits that apply for other file types.
ANNOTATED EXAMPLES:
OPEN 1,8,5," BUNCH ADAT A , PRG , R "
Opens file# 1 to disk (eight) using channel# 5, to access the file on drive zero (default)
named BUNCHADATA. It is a program (PRG) file and we want to read (R) it.
OPEN 3,8,9, "PILEAJUNK,P"
Opens file# 3 to disk (eight) using channels 9, to access the file on drive zero (default)
named PILE AJUNK. It is a program (PRG) file and we want to read it (defaults to
ID-
INPUT "NAME TO USE FOR NEW PROGRAM FILE";F*
OPEN 6,8,4,F* + ",P,W"
131
Prepares to write (W) a new program (P) file with the name contained in the variable
F$. Note how to combine the file name (F$), the file type (P) and the mode (W).
Replacing an Existing File
If you want to replace an existing program file with a new version, you can use one of
two techniques. The first technique is to scratch the old file and then write out the
revised file as though it were a new program file. The second technique is to open the
program file for a replace by putting @0: in front of the file name in the OPEN
statement. In some cases this method can cause serious problems (see WARNING in
Section 6.6). The examples below illustrate these two methods:
1. Scratch old file and then open a new program file.
250 OPEN 15,8,15:REM
OPEN COMMAND CHANNEL
260 PR I NT# 1 5 . " SO : BUNCHA JUNK " : REM
SCRATCH OLD FILE
270 INPUT#15,E,E*,T,S:REM
READ ERROR CHANNEL
280 IF E<20 GOTO 310: REM
CHECK FOR DOS ERROR
290 print e;e*;t;S:REM
PRINT ERROR MESSAGE
300 CLOSE 15: STOP: REM
ABORT ON ERROR
310 OPEN 3,8,5, "BUNCHAJUNK,P,W": REM
OPEN FILE
320 REM PROGRAM CONTINUES...
2. Open file for a replace.
250 OPEN 3,8,5, "@0: BUNCHA JUNK, P,W"
NOTE:
Be sure to read the warnings about scratching or replacing existing files in Section
6.6 before attempting this.
Writing to a Program File
Once you have opened the file in write (W) mode, use a PRINT# statement to write
information into the file. In this example, we begin by opening a new program file.
Then we print in the load address followed by the main body of the file. Finally, we
close the file.
200 REM OPEN FILE
210 OPEN 7,8, 12, "FAKE PGM,P,W"
220 :
230 REM PRINT LOAD ADDRESS INTO FILE
240 PRINT#7,CHR*(1);:REM NOTE ';' AT END
OF LINE!
250 PRINT#7,CHR*<8>;
132
260 :
270 REM PRINT OUT THE FILE
280 FOR K=l TO 85: READ X
290 PRINT#7,CHR*<X>;
300 NEXT
310 CLOSE 7
320 DATA 15,8, ... BASIC program to go into the file.
WARNING:
Be careful to always put a semicolon at the end of each PRINT# statement to suppress
the printing of the carriage return character, CHR$(13). If you forget to do this, you
will be in trouble when you try to LOAD the file.
NOTE:
The MAKE FAST COPY program listed in Appendix E is a practical example of
how to write a program file.
Reading a Program File
Because a BASIC or machine language file is not broken up into records, you must
read the file one character at a time using a GET# statement. Here is a short illustra-
tion of how to read a program file.
******************
PROGRAM: LOAD ADDR
******************
0 REM 1541 USER'S GUIDE SECTION 6.8
1 REM COPYRIGHT: G. NEUFELD, 1984
2 :
10 REM READ LOAD ADDRESS OF PROGRAM FILE
20 INPUT "NAME OF FILE";F*
30 OPEN 15,8,15 Note how the file name, file
40 OPEN 5 , 8 , 4 , F*+ " , P , R " «— type & mode are concatenated.
50 REM GET LOAD ADDRESS
60 GET#5,A*:IF A*="" THEN A*=CHR*(0)
70 LO=ASC<A*>
80 GET#5,A*:IF A*=M" THEN A*=CHR*(0)
90 HI=ASC(A*)
lOO PRINT "THE LOAD ADDRESS OF "F*" IS"
256*HI + LO
HO CLOSEl:CLOSE15:END
REMINDER:
The GET# statement does not handle nulls correctly. When you use GET# (as in
GET#5,X$) to get a $00 byte from a file, the string variable (X$) is not given the value
CHR$(0). It becomes the null string (contains no characters). If you try to find the
133
ASCII value of a null string, you will get an ?ILLEGAL QUANTITY ERROR. That
is the reason for the awkward statement
SET#5,A*:IF A*=MM THEN A*=CHR*<0>
NOTE:
The PRG HEX DUMP and PRG ANALYZER programs listed in Appendix E are
practical examples of how to read and interpret a program file.
Closing a Program File
Once you have finished using a program file, you must close it A CLOSE statement
looks like this:
SYNTAX:
CLOSE file#
examples:
CLOSE 1
CLOSE 5: CLOSE 15
WHERE:
fite# = the logical file number you used when you opened the file.
NOTE:
Closing the file is particularly important when you have written to a file. Unless you
close the file, the information you have sent to the disk drive may not even be
recorded onto the diskette. See Section 11.3 for information about how to recover
data from an unclosed file.
6.9 The Directory as a Program File
The diskette directory is a very special form of program file. Its name is $. It can be
opened and read just like any other program file. However, you cannot open this file
in write mode.
You can open this file in two different ways. You can open it using a normal data
channel number (2-14) or with the reserved channel number zero. Surprisingly, the
two different methods do not produce the same result! You may want to type in these
short programs and see for yourself.
METHOD #1: Using a normal data channel.
******************
PROGRAM: DIR CH# 6
****•♦************#
0 REM 1541 USER'S SUIDE SECTION 6.9
1 REM COPYRIGHT: 6. NEUFELD, 1984
2 :
3 REM READ DIRECTORY USING DATA CHANNEL
134
4 :
lO OPEN 1,8,6, "*,P,R"
20 K=K+I
30 GET* 1, A*: IF A*=""THEN A*=CHR*(0)
40 A=ASC(A«>
50 PRINT k;a;
60 IF A>31 AND A<97 THEN PRINT A*;
70 PRINT,
80 IF ST=0 GOTO 20
90 CLOSE 1
This method allows you to read the entire directory file one byte at a time. The first
254 bytes in this file are the BAM, the diskette name, etc. from track 18, sector 0. The
bytes that follow are from the directory entries. This method allows you to see all the
bytes in the directory: the track and sector link, and the record size and side sector
link for a relative file. It also allows you to see the file name, length of the file, and the
file type that you see in a normal directory listing.
METHOD #2: Using data channel zero.
******************
PROGRAM: DIR CH# O
******************
0 REM 1541 USER'S GUIDE SECTION 6.9
1 REM COPYRIGHT: G. NEUFELD, 1984
2 :
3 REM READ DIRECTORY USING DATA CHANNEL
4 :
10 OPEN 1,8,0,"*"
20 K=K+1
30 GET#1,A*:IF A*=""THEN A*=CHR*<0)
40 A=ASC(A«>
50 PRINT k;a;
60 IF A>31 AND A<97 THEN PRINT A*;
70 PRINT,
80 IF ST=0 GOTO 20
90 CLOSE 1
This method produces different results. Because you have used the special channel#;0
that is normally reserved for loading programs, the disk drive thinks you are prepar-
ing to LOAD the directory. As a result, the drive assembles the directory into the
pseudo program (load address $0401) form that you get when you LOAD the directory.
In this form you only get the diskette name, cosmetic diskette ID, the lengths of the
file, the file names and the file type that you see in a normal directory listing.
The second technique is handy if you want your program to be able to call up and
display a directory quickly. The two sample programs listed below illustrate how to
135
do this. The first program displays the directory listing in the same form you would
see if you loaded the directory and listed it. The second program suppresses the file
lengths and the file types. Only the file names are displayed.
*****■*♦•**»******
PROGRAM: ALL DIR
Displays directory as it appears in a normal listing.
0 REM 1541 USER'S GUIDE SECTION 6.9
1 REM COPYRIGHT: G. NEUFELD, 1984
2 :
3 REM READ AND DISPLAY ALL OF DIRECTORY
4 :
lOO REM ALL DIR
110 OPEN 15,8,15
120 OPEN 1,8,0, "*0"
1 30 GET* 1 , A* : GET# 1 , A* : BY=26
140 GET#1,A$:GET#1,A$
1 50 GET* 1 , A* : BL= ASC ( A*+CHR* ( O ) )
1 60 GET* 1 , A* : BL=BL+256*ASC < A*+CHR* < O ) )
170 N*="":FOR K=l TO BY
180 GET*1,A*:IFSTOO GOTO 220
190 N*=N*+A*:NEXT
200 printbl;n*:by=28
210 IF ST=0 GOTO 140
220 printbl;n*
230 CLOSE l: CLOSE 15
**#***♦*•*****♦****
PROGRAM: NAMES DIR
******************
Displays only the file names from the directory.
0 REM 1541 USER'S GUIDE SECTION 6.9
1 REM COPYRIGHT: G. NEUFELD, 1984
2 :
3 REM READ AND DISPLAY NAMES FROM DIRECT
ORY
4 :
100 REM NAMES DIR
HO OPEN 15,8,15
120 OPEN 1,8,0, "*0"
130 F0RK=1T030:GET#1.A*:NEXT
140 GET*1,A*:IF STOO GOTO 210
150 IF A*OCHR*(34)GOTO 140
160 GET#1,A*:IF STOO GOTO 210
170 IF A*<>CHR*(34>THENPRINTA*;:GOTO 160
136
180 N*="":F0R k=i to by
190 PRINT: G0T0140
200 n*=n*+a*:next
210 CLOSE l: CLOSE 15: END
In some cases you might like a more selective listing of the directory entries. You
could use the brute force method; read all the directory entries into an array and then
just display those that you want However, there is a much easier way, just ask for a
selective listing in your OPEN statement using wild cards. A request for a selective
listing looks like this:
SYNTAX:
OPEN 1, 8, 0,"*0: filename"
EXAMPLES:
OPEN 1,8,0,"*0:PR*"
OPEN 1,8,0, h«o:d»"
ANNOTATED EXAMPLE:
OPEN 1,8,0,"*0:D*"
The directory listing will consist of the diskette name and ID line, the entries for the
files whose file names start with the letter D, and the BLOCKS FREE line.
NOTE:
For a practical example of this technique see the MOD ENTRY program in Section
13.8.
6.10 Self-modifying Programs
Programs that modify themselves have always fascinated computer buffs. Let's look
at two different ways of doing this: direct POKEs into memory and automated editing.
Direct POKEs into Memory
The most obvious way to have a program modify itself is to have it POKE different
values into that part of RAM memory that contains the program. For example, here
is a very simple math drill program that becomes either an addition drill or a
subtraction drill depending on your response to the input in line 10.
PROGRAM: ADD/SUBT
10 INPUT "ADD OR SUBTRACT <A/S)";A*
20 I F A*= " A " THENP0KE2 1 96 , 43 : P0KE22 1 8 , 1 70
30 I F A*= "S" THENP0KE2 1 96 , 45 : POKE22 18,171
40 Nl=INTUO*RND<l>>
50 N2=INT(10*RND<1>)
60 PRINT"WHAT IS"N1 "+"N2;
137
70 INPUTA
80 AN=N1+N2
90 IFA=ANTHENPRINT"GOOD WORK"
lOO IFAOANTHENPRINT"I BET "AN "AS AN ANSW
ER"
HO GOT040
WARNING:
Do notmake any changes in lines 10 to 80. Type in this ADD/SUBT DRILL program
exactly as it appears above. Adding or subtracting even one character will make the
program inoperative. Save a copy before you run it!
NOTE:
This program modifies line 60 by POKEing either a 43 (ASCII +) or 45 (ASCII -) into
memory location 2196. It also modifies line 80 by POKEing either a 170 (+ token) or
171 (-token) into memory location 2218. The exact memory locations to use were
determined by examining a hex dump of the program. If you are using a VIC, use
locations 4244 and 4266.
If you want to add information to a program, such as adding new names to a mailing
list, you would include some dummy DATA statements like this in your program:
2000 data:::
2010 data:::
2020 data:::
As more names were added, they would be POKEd into the appropriate memory
locations and would replace the colons in these DATA statements.
The problem with this approach is that you must never add or delete characters or
lines in your program. If you do, you will end up altering line numbers, line links or
other important elements of your program when you POKE new information into
memory. The only real benefit of this approach is that the changes that are made are
internal to your program and do not affect the values of any variables that you
are using.
Automated Editing
We all know how to edit a program. You just type out the line you want to add or
modify on the screen and then press the RETURN key. What we are going to do is
have the program edit itself. The basic process is rather simple. Here is what you
have your program do:
1. Save the values of any important variables by POKEing them into a safe area of
memory (e.g., cassette buffer 828-1019).
2. Clear the screen and print out the new program line, complete with line number,
on line three of the screen.
138
3. Print RUN or GOTO xxx on the next line of the screen (xxx is the line number in
your program where the rerun should start).
4. HOME the cursor and suppress a carriage return.
5. POKE two (up to 10 if needed) carriage return characters (ASCII 13) into the
keyboard buffer (631 to 640). You need one for each new line plus one for the RUN
or GOTO. The first one goes into 631, the second into 632, and so on.
6. POKE memory location 198 with the number of characters you have stuffed into
the keyboard buffer and END the program.
7. On re-entry peek out the variable values.
This may sound a bit complicated but let's follow it through with an example. Here is
a very simple self-editing program:
******************
PROGRAM: SELF-EDIT
******************
0 REM 1541 USER'S GUIDE SECTION 6.10
1 REM COPYRIGHT: G. NEUFELD, 1984
2 :
3 REM SELF-EDIT FOR THE COMMODORE 64
4 :
100 INPUT "NEW LINE"; A* Input the new line.
HO PRINT" <CLR>CDOWN 3>"A* Clear screen & print line.
120 PR I NT "RUN" Print RUN so program will rerun.
130 PRINT" CHOME>" Home the cursor.
140 POKE63 1,13: POKE632 , 1 3 POKE returns into buffer.
150 POKE 198, 2 POKE # of characters in buffer.
160 END End the program.
Suppose you input the line 40 LIST in response to line 100. Just before the program
ends, the top of the video screen will look like this:
line 1
line 2
line 3
line 4
| <— cursor
40 LIST
RUN
As soon as the program ends, the computer will print READY and be in immediate
mode with the cursor flashing at the start of the line 40 LIST (you can see this if you
add a line 135 END to the program). However, there are two carriage return char-
acters in the keyboard buffer. The first one edits the program by entering the new
139
line 40 just as though you had pressed the RETURN key. The second character forces
a new RUN of the program. The mailing list program in the next section is a much
more useful and sophisticated application of this technique.
6.11 Typical Application
For each of the chapters on file handling (Chapters 6, 7, 8 and 9), I have developed a
mailing list program to illustrate a practical use for the techniques you have learned.
Only those portions of the program that deal with data handling are changed from
one chapter to the next. The portions relating to data entry, editing and printouts
remain the same.
In this chapter, the mailing list program is a self-modifying program that stores the
data about the club members in DATA statements.
Background information: You've just been elected secretary of a local civic organiz-
ation. As secretary, you are responsible for making up mailing labels for the monthly
newsletter that goes out to all 500 members. You would like to use your new computer
to take some of the drudgery out of the task.
Analysis of the problem: You need a simple mailing list program to keep track of the
names and addresses of the 500 members. The program should allow you to add new
members, modify members' names and addresses, produce a list of members, and
produce a set of mailing labels. Since you are interested in self-modifying programs,
you've decided to try and write one.
Planning a record: The first step in planning the program is to decide what infor-
mation we need about each member. You've checked over the club records and made
up the following summary.
Field
Description
Typical Case
Worst Case
1
Last name
8 characters
14 characters
2
First name
6 characters
9 characters
3
Street address
18 characters
23 characters
4
City
8 characters
18 characters
5
State
2 characters
2 characters
6
Zip code
6 characters
6 characters
Total
48 characters
72 characters
Data Statements: Since we are going to put the members' information into DATA
statements in our program, we must decide how many statements to use for each
member. We will enclose each field in quotation marks so that we can use both upper
and lower case letters. This means that we will have a lot of extra characters to deal
with. As a result, we had better use two DATA statements per member. Well organize
them like this:
5000 DATA"Last name", "First name", "Street address"
5005 DATA"City", "State", "Zip code"
Let's go over the main parts of the program to see how it works:
140
Line Range Description
1020-1080 Loop to read and count the number of existing data items. When -999
is encountered, exit the loop and divide by six, the number of fields
per member, to find the number of existing records. Convert this
number into lo byte/hi byte form and POKE it into memory locations
1022 and 1023.
1120-1140 Reentry point after auto editing. PEEK memory locations 1022
and 1023 to recover the number of existing records (NR) and initialize
variables.
1150-1250 Display main menu, input option chosen and branch to the appropriate
part of the program.
1290-1460 Increment count of existing members, including the values in 1022/23.
Determine line number for the new DATA lines. Input the data and
assemble the statements in Fl$ and F2$ and check their length. If
the lines are too long, exit to edit routine.
1490-1550 Clear screen, print new DATA statements and GOTO 1120 on the
screen, stuff the keyboard buffer and end the program. After the new
lines are added, reenter at 1120.
1590-1710 Input the number of the record to edit and then position to the record.
1720-1850 Read and display the existing record. Input any changes to the record.
1870-1920 Calculate the line numbers for the revised DATA statements,
assemble the lines and check length.
1940-1970 Clear screen, print revised DATA statements and GOTO 1120 on the
screen, stuff keyboard buffer and end program. After the lines are
edited, reenter at 1120.
2040-2190 Read the DATA statements and print a members list on the screen or
printer.
2230-2540 Read the DATA statements and print mailing labels to the screen or
printer. Pressing Q causes the printing to be suspended after the next
label is printed and the user has the option of aborting the print job or
restarting it at any point. This is very helpful when the paper jams!
2580-2590 Saves a new, revised copy of the program onto diskette and terminates
the program.
5000-9995 DATA statements containing names and addresses.
10000 DATA statement with -999 as end of file indicator.
141
*****************
program: mail prg
*****************
0 REM 1541 USER'S GUIDE SECTION 6.11
1 REM COPYRIGHT: G. NEUFELD, 1984
2 :
3 REM SELF-MODIFYING, MAIL-LIST PROGRAM
4 :
1000 REM* MAILING LIST PROGRAM
1010 :
1020 REM FIND NUMBER OF DATA ITEMS
1030 K=0
1040 K=K+l:READA*:IF A*=" -999 "GOTO 1060
1050 GOTO 1040
1060 NR=INT(K/6):REM 6 FIELDS PER PERSON
1070 RH=INT(NR/256):RL=NR-256*RH
1080 POKE 1023,RH:P0KE1022,RL
1090 :
1100 REM* DISPLAY MENU
1110 :
1 1 20 RH=PEEK ( 1 023 ) : RL=PEEK < 1 022 ) : NR=RL+2
56*RH
1 1 30 C*=CHR* ( 34 ) +CHR* ( 44 ) +CHR* < 34 > : REM "
II
1140 Q*=CHR*(34):REM "
1150 PRINTCHR*(14>"{CLR}{DOWN} MAILING L
1ST MENU "NR" RECORDS"
1160 PRINT" CDOWNM. ENTER NEW MEMBER"
1170 PRINT" {DOWN} 2. EDIT EXISTING RECORD
ll
1180 PRINT" {DOWN} 3. PRINT MEMBERS LIST"
1190 PRINT" CDOWN> 4. PRINT MAILING LABELS
ll
1200 PRINT" {DOWN} 5. TERMINATE PROGRAM"
1210 PRINT" { DOWN > WHICH OPTION <l-5>"
1220 GET A*: IF A*=""GOTO 1220
1230 A=ASC(A*)-48
1240 IF A<1 OR A>5 GOTO 1220
1250 ON A GOTO 1290,1590,2040,2230,2550
1260 :
1270 REM: ENTER NEW MEMBER
1280 :
1 290 RH=PEEK < 1 023 ) : RL=PEEK < 1 022 )
1300 RL=RL+l:IFRL>255 THEN RL=RL-256:RH=
RH+1
1310 POKE 1023,RH: POKE 1 022, RL
1320 NR=RL+256*RH
1330 L1*=MID*(STR*(4990+10*NR),2):REM LI
NE NUMBER
142
1340 L2*=MID*<STR*< 4995+ 10*NR>,2>: REM LI
NE NUMBER
1350 PRINT" {CLR} {DOWN} ENTER DATA FOR M
EMBER #"NR
1 360 NL*= " ?" : I NPUT " {DOWN} LAST NAME : " ; NL*
1370 NF*="?": INPUT" {DOWN} FIRST NAME:";NF
*
1380 AD*="7": INPUT" {DOWN} STREET ADDRESS:
" ; AD*
1390 ci*="?": input" <:down}city/town: m;ci*
1400 SA*="7" : INPUT" {DOWN} STATE: " ; SA*
1410 ZP»= " ? " : I NPUT " {DOWN} ZIP CODE : " ; ZP*
1 420 F 1 *=L 1 *+ " DATA " +Q*+NL*+C*+NF*+C*+AD*
+Q*
1 430 F2»=L2*+" DATA " +Q*+C I *+C*+SA*+C*+ZP*
+Q*
1440 IF LEN<F1*)<79 AND LEN<F2*)<79 GOTO
1490
1450 PRINT" {DOWN} ERROR: RECORD IS TOO LO
NG":X=NR
1460 FOR K=l TO 5000:NEXT:B0T0 1730
1470 :
1480 REM PRINT RECORD ON SCREEN
1490 PRINT" {CLR} {DOWN 3} "F1*:PRINTF2*
1500 PRINT"SOTO 1120{H0ME}"
1510 :
1520 REM STUFF KEYBORD BUFFER WITH CARRI
AGE RETURNS
1530 POKE 631, 13: POKE 632, 13: POKE 633,13
1540 POKE 198,3
1550 END
1560 :
1570 REM* EDIT EXISTING MEMBER
1580 :
1590 PRINT" {CLR} {DOWN} EDIT EXISTING MEM
BER{DOWN}"
1600 M*=" RECORD NUMBER (l-"+MID*(STR*(NR
>,2>+">"
1610 PRINTM*;" O"
1620 PRINT"{UP}"TAB(LEN(M*>);: INPUT X*:X
=INT(VAL<X*>)
1630 IF X<1 OR X>NR GOTO 1120
1640 :
1650 REM POSITION TO RECORD #X
1660 :
1670 RESTORE
1680 IF X=l GOTO 1720
143
1690 FOR K=1T0X-1
1700 READ NL*,NF*,AD*,CI*,SA*,ZP*
1710 NEXT
1720 READ NL*,NF*,AD*,CI*,SA*,ZP*
1730 PRINT" {CLR} {DOWN} EXISTING DATA FOR
MEMBER #"X
1740 PRINT" { DOWN } LAST NAME: ";NL*
1750 PRINT" {DOWN* FIRST NAME: ";NF*
1760 PRINT" {DOWNJSTREET ADDRESS: ";AD*
1770 PRINT" {DOWN}CITY/TOWN: ";ci*
1780 PRINT" {DOWN} STATE: ";SA*
1790 PRINT" {DOWN* ZIP CODE: ";ZP*
1800 PRINT" {HOME* {DOWN 3} "TAB (10) : INPUT
NL*
1810 PRINT"{D0WN}"TAB<11>: INPUT NF*
1820 PRINT" {DOWN} "TAB (15): INPUT AD*
1830 PRINT "{DOWN}" TAB <10): INPUT CI*
1840 PRINT" {DOWN} "TAB <6>: INPUT SA*
1850 PRINT" {DOWN} "TAB (9): INPUT ZP*
1860 REM FIND LINE NUMBER
1870 L1*=MID*(STR*<4990+10*X),2):REM LIN
E NUMBER
1880 L2*=MID*(STR*<4995+10*X),2):REM LIN
E NUMBER
1890 REM ASSEMBLE LINE
1 900 F 1 *=L 1 *+ " DATA " +Q*+NL*+C*+NF*+C*+ AD*
+Q*
1910 F2*=L2*+ ■ DATA " +Q*+C I *+C*+SA*+C*+ZP*
+Q*
1920 IF LEN(F1*)>79 OR LEN(F2*>>79 GOTO
1730
1930 REM PRINT RECORD ON SCREEN
1940 PRINT" {CLR} {DOWN 3}"F1*:PRINTF2*
1950 PR I NT "GOTO 1 120 {HOME}"
1960 :
1970 REM STUFF KEYBOARD BUFFER WITH CARR
I AGE RETURNS
1980 POKE 631, 13: POKE 632, 13: POKE 633,13
1990 POKE 198,3
2000 END
2010 :
2020 REM* PRINT LIST OF MEMBERS
2030 :
2040 PRINT" {CLR} {DOWN} PRINT LIST OF MEM
BERS"
2050 INPUT" {DOWN} OUTPUT TO SCREEN OR PR I
NTER (S/P) S{LEFT 3}";Z*
2060 PRINT:DN=3:IF LEFT*<Z*, 1)="P" THEN
DN=4
144
2070 0PEN4,DN: RESTORE
2080 FOR K=l TO NR
2090 REM POSITION TO RECORD #K
2100 READ NL*,NF*,AD*,CI*,SA*,ZP*
2110 PRINT#4, RIGHT* <" "+STR* (K) .3) ; " "
NF*" "NL*
2120 PRINT#4," "AD*", "CI*", "SA*% "
ZP*
21 30 NEXT
2140 CLOSE4
2150 IF DN=4 GOTO 1120
2160 PRINT" <DOWN> PRESS CRVS>RETURN£ROF
F> FOR MENU"
2170 GET A*: IF A*<>"" GOTO 2150
2180 GET A*: IF A*OCHR*<13) GOTO 2180
2190 GOTO 1120
2200 :
22 lO REM* PRINT MAILING LABELS
2220 :
2230 PRINT" {CLRXDOWN* PRINT MAILING LAB
ELS"
2240 INPUT" {DOWN} OUTPUT TO SCREEN OR PR I
NTER (S/P> SO.EFT 3>";Z*
2250 PRINT" {DOWN} PRESS Q TO SUSPEND PR IN
TING"
2260 PRINT:DN=3:IF LEFT*<Z*, 1>="P" THEN
DN=4
2270 0PEN4,DN: RESTORE
2280 FOR K=l TO NR
2290 REM POSITION TO RECORD #X
2300 READ NL*,NF*, AD*,CI*,SA*, ZP*
23 lO PRINT#4,NF*" "NL*","
2320 PRINT#4,AD*","
2330 PRINT#4,CI*", "SA*", "ZP*
2340 PRINT#4:PRINT#4
2350 GET A*: IF A*="Q" GOTO 2440
2360 NEXT
2370 CL0SE4
2380 IF DN=4 GOTO 1120
2390 PRINT" {DOWN} PRESS <RVS} RETURN CROF
F} FOR MENU"
2400 GET A*: IF A*<>"" GOTO 2400
2410 GET A*: IF A*OCHR*<13) GOTO 2410
2420 GOTO 1120
2430 :
2440 PRINT" {DOWN}OUTPUT ABORTED"
2450 PRINT" {DOWN} COMMANDS: A = ABORT"
2460 PRINT" R = RESTART"
2470 GET A*: IF A*="A"THEN CL0SE4:G0T0 11
20
145
2480 IF A*<>"R" GOTO 2470
2490 PRINT" { DOWN > RECORD #"K"HAS JUST BEE
N PRINTED"
2500 PRINT" <DOWN> RESTART LIST AT RECORD
# "K+l
2510 PRINT" tUP> "TAB (24): INPUT K:PRINT
2520 RESTORE: IF K=l GOTO 2540
2530 FORL=lTOK-l:READ NL*,NF*, AD*, CI*, S A
*,2P*:NEXT
2540 GOTO 2380
2550 :
2560 REM* TERMINATE PROGRAM
2570 :
2580 SAVE "SO: MAIL PRG",8
2590 PRINT" {CLRXDOWNJ REVISED PROGRAM S
AVED"
5000 DATA" CSHIFT-SJMITH" , " tSHIFT-R>OBERT
","125 €SHIFT-M>AIN <SHIFT-S>TREET"
5005 DATA" £SHIFT-A>NYTOWN" , " CSHIFT-N>EW
<SHIFT-Y>ORK", "12345"
5010 DATA" tSHIFT-J>ONES" , " CSHIFT-S>ALLY"
,"127 CSHIFT-FMRST CSHIFT-S>T. CSHIFT-N
><SHIFT-W>"
5015 DATA" <SHIFT-N>ORTHLAND" , " <SHIFT-M> I
SHIFT-N>", "56024"
lOOOO DATA -999: REM DATA END FLAG
Possible modifications and improvements:
1. Have the new DATA lines and GOTO 1120 printed on the screen in the same color
as the background. This way they would be less obvious.
2. Have the names sorted alphabetically. To do this you would have to devise a
different method of determining the line numbers for the DATA statements. The
line number would have some mathematical relationship to the characters that
made up the name (this is called hashing).
146
Chapter
7
SEQUENTIAL FILES
In this chapter we will take a close look at sequential files, their structure and use.
Several unique applications, such as using sequential files for merging program
segments and hiding programs on a disk, are presented.
7.1 Introduction to Sequential Files
A sequential file is the most common type of data file. It is easy to use and makes very
efficient use of diskette storage. Each record is just long enough to hold the information
stored there. This type of file is particularly useful for storing relatively small amounts
of data, up to about 20K. With this amount of data you can read all the data into the
computer's memory, modify it in any way you want, and then rewrite the new version
of the file. With a single disk drive, you can use a sequential file to hold up to a
maximum of about 85K characters (half a disk full). But, if you use this large a file,
you will find that updating or editing the data becomes quite tricky.
The biggest limitation of a sequential file is that you must access the file in a sequential
manner. For example, if you want to read the address of the 100th person in your
mailing list, you have to read through the first 99 records to get to the record you
want. Another limitation is that you cannot modify individual records. You must read
in the data, modify it and then rewrite the entire file, even if all you wanted to do was
change one character in one record.
When to use a sequential file:
1. If you have a relatively small amount of data to store.
2. If you have up to 80K of data that rarely needs to be updated or modified.
The benefits from using sequential files are:
1. They are easy to use.
2. The data is stored very compactly.
The drawbacks in using sequential files are:
1. The amount of data you can store in a sequential file is usually limited to the
amount of data that can be held in the computer's memory at one time.
147
2. You must read through all the preceding records to read a particular record.
3. You cannot modify just one record in the file, you must read in the whole file,
modify the record(s), and then rewrite the entire file.
Some typical uses of sequential files are:
1. Grade book programs.
2. Phone book programs.
3. ASCII listings of programs that are to be merged (used by the programmer's aid
program POWER™).
4. The management of small data bases.
5. As an index into a large relative file *.
* The use of a sequential file as an index into a large relative file is a very important
and powerful technique. Chapter 9 on indexed relative files shows you how to do this.
7.2 Structure of a Sequential File
A sequential file is usually composed of many records. Each record stores all the
information about a single person, place or thing. Since the length of a record is not
fixed, the records are usually of varying lengths. A typical file might look like this:
Record #1
Record #2
Record #3
Record #4
Since each record occupies only as much space as it needs, the information is packed
as tightly as possible on the diskette. This makes for efficient use of diskette storage
space. However, it also means that there is no easy way to determine exactly where
any given record starts or stops. The only way to read the Nth record is to start
reading at the beginning of the file and read through N-l records. The variable
record size also makes revising a record difficult. You cannot rewrite an individual
record in a sequential file. You must rewrite the entire file to revise a single record.
7.3 Using a Sequential File
Creating a program that reads or writes a sequential file is quite easy. Just follow
these steps:
1. OPEN the sequential file.
2. Write information into the file using PRINT#.
3. Read information from the file using INPUT# or GET#.
4. CLOSE the sequential file when you are finished.
Let's look at each of these steps in more detail:
Opening a Sequential File
(See also Sections 7.4 and 7.5.)
148
An OPEN command is used to tell the disk drive the name of the file you wish to use.
On the Commodore 64 or the VIC-20 the OPEN command to access a sequential file
looks like this:
syntax:
OPEN f i 1 e#, devi ce#, channel #, "f i 1 ename, SEQ, mode"
EXAMPLES:
OPEN 5,8,4,"MEMBERS,SEQ,R"
OPEN 1,8,5," BUNCH AD AT A , S , R "
F*=" SUPER FILE"
OPEN 2,8, 12,F*+",S,W"
WHERE:
file# = the logical file number (any integer 1-127).
channel# = the channel number (any integer 2-14).
filename = the file name (up to 16 characters).
SEQ = file type, may be abbreviated with S.
mode = R or W to indicate whether you want to read from a file or
write to a file.
NOTES:
1. Part or all of the file identification information may be stored in a string variable.
2. The file type and mode are optional. If the file type is omitted, the file type will be
determined from the directory entry for the file. If the mode is omitted, it defaults
to READ mode.
3. There is a special MODIFY (M) mode for reading an unclosed file. See Section
11.3 for more details.
READ MODE:
4. The file name may contain one or more wild cards (? or *). If it does, the first file
name that matches the pattern will be opened.
5. If there is no file with the name specified stored on the diskette, you will get a 63,
FILE NOT FOUND, error message.
WRITE MODE:
6. The file name may not contain any wild cards (? or *).
7. If there already is a file with the specified name stored on the diskette, you will get
a 62, FILE EXISTS, error message. To write over (replace) an existing file see
Section 7.4.
149
REMINDERS:
1. Monitor the error status of the disk drive.
You should use the command channel to monitor the error status of the disk drive.
Be sure to check the status whenever you open a file to make sure that the file has
been opened properly. If the file is not open, your new data will be lost! See Section
5.10 about how to do this.
2. OPEN the command channel first and close it last.
Do not open and close the command channel each time you want to check the error
status. Open the command channel at the beginning of your program and leave it
open until the end. When you CLOSE the command channel all open files are
closed automatically*.
3. All open files must have different logical file numbers.
Since the logical file number is used to identify one particular file, you must use
different logical file numbers for each of the files that are open simultaneously.
Once you close a file, you are free to reuse that logical file number in your program.
4. All open files must have different channel numbers.
Since the channel number (secondary address) identifies a particular data commun-
ication channel in the disk drive, each file that is open must have a unique channel
number. Once you close a file, you are free to reuse that channel number in your
program.
5. You can only have a few files open at once.
Each sequential or program file that you have open takes up one of the buffers in
the disk drive. Since the disk has only four buffers and one is reserved for internal
use, you can only have three files open at once (not five as indicated on page 45 of
your 1541 manual)! See Section 5.11 for the limits that apply for other file types.
ANNOTATED EXAMPLES:
35 OPEN 1,8,5, "OUT ADATE,SEQ, READ"
Opens file# 1 to disk (eight) using channel# 5 to access the file on drive zero (default)
with the name OUTADATE. It is a sequential (SEQ) file and we want to read
(READ) it.
110 OPEN 3,8, 12, "O: MISHMASH, S,R"
Opens f ile# 3 to disk (eight) using channel# 12 to access the file on drive zero with the
name MISHMASH. This is a sequential (S) file and we want to read (R) it.
10 OPEN 9,8,6, "PILEAJUNK"
Opens file# 9 to disk (eight) using channel# 6 to access the file on drive zero (default)
with the name PILEAJUNK. If it is a sequential, program or user file, it will be
opened in READ mode. If it is a relative file, it will be opened in READ/WRITE
mode.
100 F*=" CONFUSION"
110 OPEN 7,8,2, "0:"+F*+",S,W"
150
Opens file# 7 to disk (eight) using channel# 2 to create a file on drive zero with the
name CONFUSION. This is a new sequential (S) file and we want to write (W) to it.
Note how to combine the drive#, file name, file type and mode using concatenation (+).
1 00 NL*= " O : I NCONSEQUENT I AL , SEQ , WR I TE "
110 OPEN 2,8,5,NL*
Opens file# 2 to disk (eight) using channel* 5 to create a file on drive zero with the
name INCONSEQUENTIAL. This is a new sequential (SEQ) file and we want to
write (WRITE) it.
Writing to a Sequential File
Once you have opened a sequential file in write mode, you use a PRINTS statement to
write information into the file. Be sure to plan the records so that they can be read
back easily. I strongly recommend that you use carriage return characters as both
field and record delimiters.
Here is a brief program that illustrates how to input data and how to write it to
diskette immediately. Carriage return characters are used as both field and record
delimiters:
it-*******************
PROGRAM: PHONE/IMMED
0 REM 1541 USER'S GUIDE SECTION 7.3
1 REM COPYRIGHT: G. NEUFELD, 1984
2 :
3 REM WRITE AS SOON AS ENTERED
4 :
10 OPEN 15,8,15
20 OPEN 1,8,5, "SO: PHONE LIST,S,W"
30 INPUT#15,E,E*,T.S
40 IF E>19 THEN PRINT E;E*;T;S:CL0SE1 :CL
0SE15:ST0P
50 INPUT "ENTER NAME OR END TO QUIT";N*
60 IF N*="END" GOTO 110
70 INPUT "PHONE NUMBER ";P*
80 PRINT#1,N*
90 PRINT#1,P*
lOO GOTO 30
110 CLOSE l: CLOSE 15: END
To change the above program to use commas as field delimiters and carriage returns
as record delimiters, delete line 90 and change line 80 to read:
80 PRINT#1,N* "," P*
In many cases, it is more convenient to input all the data into an array and, once all
the data has been entered, write it to the diskette in one burst. This program illustrates
this second technique.
151
program: phone/array
»****»**»**♦******#*
0 REM 1541 USER'S GUIDE SECTION 7.3
1 REM COPYRIGHT: G. NEUFELD, 1984
2 :
3 REM STORE IN ARRAY, THEN WRITE
4 :
lO DIM N«(100),P*<100)
20 K=0
30 INPUT "ENTER NAME OR END TO QUIT";N*(
K)
40 IF N*<K>="END" GOTO 70
50 INPUT "PHONE NUMBER" ;P*<K>
60 K=K+l:lF K<101 GOTO 30
70 OPEN 15,8,15
80 OPEN 1,8,5, "SO: PHONE LIST,S,W"
90 INPUT#15,E,E*,T,S
100 IF E>19 THEN PRINT E;E*;T;S:CL0SE1 :C
L0SE15:ST0P
HO FOR L»0 TO K-l
120 PRINT#1,N*(L)
130 PRINT#1,P*(L)
140 NEXT L
ISO CLOSE l: CLOSE 15: END
This second approach is often more useful because it allows you to sort or edit the list
before the records are written into the sequential file. One disadvantage of this
approach is that all the data must fit into the computer's RAM memory.
Reading a Sequential File
Whenever you open a sequential file in read mode you are automatically positioned to
the start of the first record in the file. You may use either an INPUT# or GET#
statement to read information from the file. In general, plan to use INPUT# rather
than GET# because INPUT# reads more information at one time. However, there are
some situations when you must use GET# to examine a file closely.
When to use GET#:
1. To check the contents of a file that won't read properly.
2. To read a file with more than 88 characters between delimiters.
3. To read a file that contains stray commas, colons or null bytes that cause problems
for INPUT#.
Whether you use INPUT# or GET# you always start reading at the very beginning of
the file. As you read records, fields or individual characters from the file the disk
drive will keep track of where you are in the file. Note that you can only read
forwards in the file. You cannot back up. If you have read past a piece of data that you
need, you must close the file, reopen it and start from the beginning again.
152
There are two different approaches to reading a sequential file. One way is to read in
the data piecemeal. When you need a particular record, you read through the file
until you find it. You must use this approach when you have more data than can be
held in memory at once. The other approach is to read the entire file into one or more
arrays in memory in one burst. This is often a useful approach when your data may
need to be edited or sorted.
Here is a brief program that illustrates the first approach reading the data as it is
needed:
*******************
PROGRAM: READ/IMMED
*******************
0 REM 1541 USER'S GUIDE SECTION 7.3
1 REM COPYRIGHT: G. NEUFELD, 1984
2 :
3 REM READ DATA FILE AS NEEDED
4 :
10 REM FIND PHONE NUMBER GIVEN THE NAME
20 OPEN 15,8, 15: REM OPEN COMMAND CHANNEL
30 OPEN 1,8,5, "O: PHONE LIST": REM OPEN SE
Q FILE
40 INPUT#15,E,E*,T,S:REM CHECK DISK ERRO
R STATUS
50 IF E>19 THEN PRINT E;E*;T;S:CL0SE1 :CL
0SE15:ST0P
60 INPUT "ENTER NAME OR END TO QUIT";N*
70 IF N*="END" GOTO ISO
80 :
90 INPUT#1,NAM*,PH*
lOO IF N*=NAM* THEN PR I NT "NUMBER IS "PH*
: CLOSE l: GOTO 30
110 IF ST=0 GOTO 90: REM CHECK FOR END OF
FILE
120 :
130 PRINT "NAME NOT FOUND IN FILE"
140 :
150 CLOSE l: CLOSE 15: END
NOTE:
The status variable, ST, is used in line 110 to check for the end of the file. For more
details see Section 5.15.
Here is the same program again but this time reading all the data into arrays at the
start of the program:
153
*******************
program: read/ array
*******************
0 REM 1541 USER'S BUIDE SECTION 7.3
1 REM COPYRIGHT: G. NEUFELD, 1984
2 :
3 REM READ DATA FILE INTO ARRAY
4 :
lO REM FIND PHONE NUMBER GIVEN THE NAME
20 DIM NAM*<100),PH*(100)
30 OPEN 15,8, 15: REM OPEN COMMAND CHANNEL
40 OPEN 1,8,5, "O: PHONE LIST":REM OPEN SE
Q FILE
50 INPUT#15,E,E*,T,S:REM CHECK DISK ERRO
R STATUS
60 IF E>19 THEN PRINT E;E*;T;S:CLOSEl:CL
0SE15:ST0P
70 K=0
80 K=K+1
90 INPUT#1,NAM*(K),PH*(K)
lOO IF ST<>64 GOTO 80: REM CHECK FOR END
OF FILE
llO CLOSE l: CLOSE 15
120 :
130 INPUT "ENTER NAME OR END TO QUIT";N*
140 IF N*="END" THEN END
150 :
160 FOR C=l TO K
170 IF N*=NAM*(C) THEN PR I NT "NUMBER IS "
PH*(C):GOTO 130
180 NEXT
190 :
200 PRINT "NAME NOT FOUND IN FILE"
210 GOTO 130
NOTES:
1. The status variable, ST, is used in line 100 to check for the end of the file. For more
details see Section 5.15.
2. The mailing list program at the end of this chapter is a good illustration of the use
of this technique.
Let's take one last look at the first part of this same program again, but this time we
will use GET# to read the sequential file.
154
******************
PROGRAM: READ/6ET#
******************
0 REM 1541 USER'S GUIDE SECTION 7.3
1 REM COPYRIGHT: G. NEUFELD. 1984
2 :
3 REM READ DATA FILE INTO ARRAY
4 :
lO REM FIND PHONE NUMBER GIVEN THE NAME
20 DIM NAM*(100>,PH*<100>
30 OPEN 15,8, 15: REM OPEN COMMAND CHANNEL
40 OPEN 1,8,5, "O: PHONE LIST":REM OPEN SE
Q FILE
50 INPUT#15,E,E*,T,S:REM CHECK DISK ERRO
R STATUS
60 IF E>19 THEN PRINT E; E*; T;S:CLOSEl:CL
OSE15:STOP
70 K=0
80 K=K+1
90 GET#1,A*:IF A*="" THEN A*=CHR*(0)
91 IF A*=CHR*<13> GOTO 95: REM CARRIAGE R
ETURN
92 NAM*(K)=NAM*(K)+A*
93 GOTO 90
94 :
95 GET#1,A*:IF A*="" THEN A*=CHR*(0)
96 IF A*=CHR*(13) GOTO 100:REM CARRIAGE
RETURN
97 PH*(K)=PH*(K)+A*
98 GOTO 95
100 IF ST<>64 GOTO 80: REM CHECK FOR END
OF FILE
110 CLOSE l: CLOSE 15
120 :
130 INPUT "ENTER NAME OR END TO QUIT";N*
140 IF N*="END" THEN END
150 :
160 FOR C=l TO K
170 IF N*=NAM*(C) THEN PR I NT "NUMBER IS "
PH*(C):GOTO 130
180 NEXT
190 :
200 PRINT "NAME NOT FOUND IN FILE"
210 GOTO 130
(lines 100-210 as before)
155
NOTES:
1. The status variable, ST, is used in line 100 to check for the end of the file. For more
details see Section 5.15.
2. When you use GET# to read the file like this, you must build up long strings using
concatenation (X$=X$+A$). This can cause problems on a VIC-20 or Commodore
64 because you end up with a lot of unused string fragments in memory. For
example, in building up the string NANCY, you create the following fragments:
N, NA, NAN and NANC. These unused fragments gradually fill up the available
RAM memory. Finally, you reach a point where the memory is full and the
computer has to sort through the mess and discard all the strings not in use. The
sorting process is called garbage collection and can take anywhere from five
seconds to 20 minutes to complete. While garbage collection is taking place, the
cursor goes away and nothing seems to be happening. If this happens to you, don't
panic. Your computer is just busy sorting out the mess you have created! Don't
turn off your machine. Take about a 30 minute coffee break. If your machine still
appears to be hung up, your program may contain an infinite loop. Press
RUN/STOP and do some snooping (you can always carry on with CONT). If this
doesn't work, you may have to use RUN/STOP and RESTORE or even power
down. Check for bad POKE statements.
Closing a Sequential File
Once you have finished using a sequential file, you must CLOSE it. Closing a file is
particularly important after you have written to the file. A close statement looks like
this:
SYNTAX:
xx CLOSE -file#
examples:
75 CLOSE 1
600 CLOSE 5: CLOSE 15
WHERE:
xx = the line number.
f ile# = the logical file number you used when you opened the file.
WARNING:
Be sure to close a file which you opened in write mode. Closing the file is particularly
important when you are writing to a file. Unless you close the file, the last block of
data will not be recorded onto the diskette and you will have an unclosed file. You will
be unable to access any of your data unless you use the techniques described in
Section 11.3. Even if you do, some of your data will be lost!
7.4 Replacing an Existing Sequential File
Normally if you use the name of an existing file when you attempt to open a sequential
file, you get a FILE EXISTS error and the file is not opened. However, there are
times when you want to replace an existing file with a revised version. This is called
156
REPLACING the file. You can use either one of two techniques. The first technique
is to scratch the old file and then write out the revised file as though it were a new
sequential file. The second technique is to Open the sequential file for a replace by
putting @0: in front of the file name in the open statement. In some cases this method
can cause serious problems (see WARNING below).
1. Scratch old file and then open a new sequential file.
250 OPEN 15,8, 15: REM
OPEN COMMAND CHANNEL
260 PRINT#15, "SO:BUNCHAJUNK":REM
SCRATCH OLD FILE
270 INPUT#15,E,E*,T,S:REM
READ ERROR CHANNEL
280 IF E<20 GOTO 310: REM
CHECK FOR DOS ERROR
290 print e;e*st;s:rem
PRINT ERROR MESSAGE
300 CLOSE 15: STOP: REM
ABORT ON ERROR
310 OPEN 3,8,5, " BUNCH AJ UNK, S,W" : REM
OPEN FILE
320 REM PROGRAM CONTINUES. . .
2. Open file for a replace.
Replacing a file is just like writing a file, except that you are replacing an existing file
with a newer version rather than creating a new file.
SYNTAX:
OPEN f ile#,device#, channel #, "@: filename,
SEQ,mode"
ALTERNATE:
OPEN file#,device#,channel#,"@drive#: filename,
SEQ,mode"
EXAMPLES:
OPEN 5, 8,4, "@: OLDSTUFF, SEQ, W"
OPEN 1,8,5, "SO: LOTTADATA,S,W"
F*=" CURRENT MONTH" Note how the different
OPEN 2, 8, 12, "@: »+F*+" , S, W" < parts may be concatenated.
WHERE:
file# = the logical file number (any integer 1-127).
channel# = the channel number (any integer 2-14).
@drive# = @ sign followed by the drive number; the drive# (zero) is
optional on the 1541.
filename = the file name (up to 16 characters); may be contained in a
string variable.
157
SEQ = file type, may be abbreviated with S.
W = write mode required for replacement.
WARNINGS:
1. Do notscr&teh or replace improperly closed files. Be sure that the original program
file you are going to replace has been properly closed. A file that has not been
properly closed has an asterisk beside its file type in the directory (*SEQ). Do not
scratch or replace an improperly closed file. Use the VALIDATE command (see
Section 4.8) to eliminate these files! If you do scratch or replace one of these files by
accident, VALIDATE the diskette as soon as possible, but certainly before writing
to it again.
2. Be sure there is enough disk space to do a replace. When the file is opened for a
replace, the old file is not deleted 'before the new file is written. This means that
you must have enough free space on the diskette for the new copy. If there is not
enough space, the replace is aborted. The new file is incomplete and the old file is
unrecoverable unless you use the techniques given in Section 11.8. Be safe, not
sorry! Check the number of blocks free on your diskette before you try a replace
(see Section 5.13).
NOTES:
1. The file name may contain one or more wild cards (? or *). If it does, the first file
name that matches the pattern will be replaced.
2. If there is no file with the name specified stored on the diskette, a new sequential
file will be created.
REMINDER:
Part or all of the file identification information may be stored in a string variable.
ANNOTATED EXAMPLES:
20 OPEN 1,8,5, "@0:NEWSTUFF, SEQ, W"
Opens file# 1 to disk (eight) using channel# 5, to replace (@) the file on drive zero
named NEWSTUFF. It is a sequential (SEQ) file and we want to write (W) it.
BO F*="TRIV*,W"
90 OPEN 3,8,7,"@0:"+F*
Opens f ile# 3 to disk (eight) using channel* 7, to replace (@) the first file on drive zero
whose name begins with TRIV. The file type is determined from the directory entry.
The file will be opened in write (W) mode.
7.5 Appending Data to a Sequential File
Normally when you open a sequential file in write mode you begin to write at the
beginning of the file and write out the entire file. However, there are times when you
want to add new information onto the end of an existing file without disturbing the
data already in the file. This is called APPENDING data to the file. To do this you
use a slightly different OPEN statement. It looks like this:
158
syntax:
OPEN fi 1 e#, devi ce#, channel #, "-f il ename, SEQ, A"
EXAMPLES:
OPEN 5,8,4," MORESTUFF , SEQ , A -
OPEN 1,8,5, "BIGBUNCH,S,A"
F*="MORE EXPENSES" Note how the different
OPEN 2 , 8 , 1 2 , F*+ " , S , A " < parts may be concatenated.
WHERE:
file# = the logical file number (any integer 1-127).
channel# = the channel number (any integer 2-14).
filename = the file name (up to 16 characters); may be contained in a
string variable.
SEQ = file type, may be abbreviated with S.
A = append mode.
WARNING:
Before you append more data to a file, check to be sure that you have enough free
space on the diskette for the expanded file. If there is not enough space, the append
job is aborted and you have produced an unclosed file. Both the old data and the new
data are unrecoverable unless you use the techniques given in Section 11.3. Be safe,
not sorry! Check the number of blocks free on your disk before you expand that file
(see Section 5.13).
NOTES:
1. The file name may contain one or more wild cards (? or *). If it does, the first file
name that matches the pattern will be opened.
2. If there is no file with the name specified stored on the diskette, you will get a 62
FILE NOT FOUND, error message.
REMINDER:
Part or all of the file identification information may be stored in a string variable.
ANNOTATED EXAMPLES:
20 OPEN 2,8,9, "O: BILLS PAID, SEQ, A"
Opens file# 2 to disk (eight) using channel# 9, to the file on drive zero named BILLS
PAID. It is a sequential (SEQ) file and we want to append (A) data to it.
80 F*="EXP»,A"
90 OPEN 9,8,3,F*
Opens file# 9 to disk (eight) using channel# 3, to the first file on drive zero (default)
whose name begins with EXP. The file type is determined from the directory entry
The file will be opened in append (A) mode.
159
Let's take a look at a short program that creates a sequential file, then appends some
more data to it. You may want to type in the example to see how appending works. In
lines 10-50 of the example, a new sequential file is created. Lines 70-110 append new
information to the end of our existing file. Lines 130-170 read back and display
the file.
*****************
PROGRAM: APPENDER
*****************
0 REM 1541 USER'S GUIDE SECTION 7.5
1 REM COPYRIGHT: G. NEUFELD, 1984
2 :
3 REM WRITE SEQ FILE, APPEND DATA, READ
BACK
4 :
10 OPEN 2,8,5, "SO: TEST FILE,S,W"
20 FOR K=l TO 15
30 PRINT#2, "ORIGINAL RECORD #";K
40 NEXT
50 CLOSE 2
60 :
70 OPEN 3,8,7, "TEST FILE, A"
80 FOR K=l TO lO
90 PR I NT#3, "APPENDED RECORD #";K
lOO NEXT
110 CLOSE 3
120 :
130 OPEN 1,8,4, "TEST FILE"
140 INPUT#1,A*
150 PRINT A*
160 IF ST=0 GOTO 140: REM CHECK FOR END-0
F-FILE
170 CLOSE 1
Append mode is very much like write mode, except this time you begin writing data
at the end of an existing file rather than at the start of a new or replacement file.
7.6 Handling Veiy Large Sequential Files
For some applications it is necessary to use very large sequential files (over 30K of
data). In these cases there is simply too much data to be read into memory at once; you
have to read it piecemeal. This makes sorting or editing very difficult, particularly
with a single disk drive like the 1541. Here are some suggestions for handling this
type of situation.
Whenever you want to edit or add to your data file you will have to work with two
sequential files: your original data file and a new modified one. You read records one
at a time from your original file and, if nothing needs to be changed, you write them
160
out immediately to the new file. When you reach a record that you want to change,
you modify it before you write it out to the new file. Diagrammatical]./ the process
looks like this:
Original file (being read)
Adams, Tom<cr > Andrews, PauKcr >BaXXr, Fred<cr>Baxter , Don<cr >.
Bad Record
This record read but not
Modified file (being written) yet written to new file.
^>
Ad ams,Tom<cr> Andrews, PauKcr >Barker,Fred<cr>.
Edited Record
t>
To add a new record you read records from the original file and write them into the
new file. When you get to the position of the new record, you write it into the new file.
Then you carry on with the read-old/write-new process. Adding a record looks like
this:
Original file (being read)
Adams, Tom<cr>Andrews, PauKcr >Barkman,Bob<cr>Baxter,Don<cr>.
^>
This record read but not
Modified file (being written) yet written to new file.
Adams , Tcwi< cr >Andr ews , Paul <cr >Bar ker , Fred< cr >Bar kman , Bob< cr >.
IHHHM«««««
Added Record
This is what is known as an insertion sort. The new record is inserted in its proper
place among the old, sorted records. If you follow this procedure whenever the file is
updated, the records will always be correctly sorted. The sample program shown
below illustrates how to insert a single record into a large file.
******************
PROGRAM: INSERTION
******************
0 REM 1541 USER'S GUIDE SECTION 7.5
1 REM COPYRIGHT: G. NEUFELD, 1984
2 :
3 REM INSERTION SORT FOR SEQUENTIAL FIL
E
4 :
161
lOO OPEN 15,8, 15: REM
OPEN COMMAND CHANNEL
110 OPEN 1,8,5, "O: OLD FILE, S,R": REM
OPEN OLD FILE
120 INPUT#15,E,E*,T,S:REM
CHECK DISK STATUS
130 OLD=l:IF E=62 THEN OLD=0: GOTO 160: REM
NO OLD FILE
140 IF E>19 GOTO 400: REM
ABORT
ISO :
160 OPEN 2,8,7,"@0:NEW FILE, S,W": REM
OPEN NEW FILE
170 INPUT#15,E,E*,T,S:REM
CHECK DISK STATUS
180 IF E>19 GOTO 400:REM
ABORT
190 INPUT "NEW LAST NAME "; NL* : REM
RECORD TO INSERT
200 INPUT "NEW FIRST NAME";NF*
210 w=o:rem
NEW RECORD WRITTEN
220 IF OLD=0 THEN GOSUB 360:G0T0 310: REM
NO OLD FILE
230 :
240 I NPUT# 1 , OL* , OF* : REM
READ OLD RECORD
250 S=ST:REM
STORE STATUS VALUE
260 IF OL*>NL* AND W=0 THEN GOSUB 360: RE
M FOUND THE SPOT
270 PR I NT#2 , OL* " , " OF* : REM
WRITE OLD RECORD
280 IF S<>64 GOTO 240:REM
MORE IN OLD FILE
290 IF W=0 THEN GOSUB 360: REM
WRITE NEW IF LAST
300 :
310 CLOSE l:CL0SE2: REM
CLOSE FILES
320 PRINT#15, "SO: OLD FILE": REM
SCRATCH OLD FILE
330 PRINT#15, "RO:OLD FILE=NEW FILE": REM
NEW BECOMES OLD
340 CL0SE15:END:REM
FINISHED
350 :
360 PR I NT#2 , NL* " , " NF* : REM
WRITE NEW RECORD
370 W=l:REM
SET FLAG
162
380 RETURN
390 :
40O printe;e*;t;s:rem
ABORT ON ERROR
410 CLOSE 1 : CL0SE2 : CLOSE 1 5 : STOP
It is inefficient to read and rewrite a large file to modify just one record as we did in
this example. You probably want to design your program so that all the changes you
want to make are input first. Once they are all input, your program would sort them
and then begin the read-old/write-new update sequence.
WARMING:
Be careful not to create too large a file. If you only have one 1541 drive, you must
never create a sequential file that occupies more than 50 percent of the diskette. If
you do, there won't be space on the diskette for both copies of the file, the old and the
new, so you will be unable to edit or update the file. If you have two drives, you can
use a complete diskette because you can always have the original in one drive and the
updated version in the other drive.
7.7 Storing a Program as a Sequential File
Most Commodore users do not know that they can store a copy of a BASIC program
in a sequential file. There are two ways to do this. One way produces an ASCII text
file of the program while the other produces a sequential program file.
ASCII Text Hies
In this type of file the BASIC keywords, such as PRINT and INPUT, are spelled out
as they are in a program listing rather than tokenized as they are in a program in
RAM memory or in a normal program file. Producing an ASCII text file of a program
is very similar to producing a program listing on a printer. Here are the commands
to enter in immediate mode:
OPEN 1,8,5, " -f i 1 en ame , S , W " Open sequential file in write mode.
CMD 1 : L I ST Divert output to file and list program.
The LIST command in the second line may be any one of the usual variations if you do
not want the entire program listed. Once the flashing cursor returns, close the file
like this:
PR I NT# 1 Cancel CMD command.
CLOSE 1 Close diskette file.
Although the file contains a copy of your program, it cannot be loaded because the
program is no longer in its tokenized form. The main use of this type of file is to
merge subroutines you have stored on diskette into your current program. The excel-
lent programmer's utility, POWER™, uses this technique.
163
HINT:
If you don't have access to POWER™, this four line BASIC routine is all you need to
merge programs.
PROGRAM: 64 MERGE
ft-****************
0 OPEN 1,8 ,5," TEST"
1 POKE 152. l: PRINT" {CLRXDOWN> "
2 GET# 1 , A* : PR I NTA* ; : I F A*< >CHR* < 1 3 ) G0T02
3 PRINT"GOT01CHOME>"; :P0KE631, 13:P0KE632
.13:POKE198.2:END
To use this routine follow these steps:
1. Check through a listing of the subroutine or program segment to be sure there are
no lines more than 80 characters long. Lines more than 80 characters must be
split into two shorter lines.
2. Save your subroutine as an ASCII file as described above.
3. LOAD your main program into memory.
4. Type in the routine as lines 0-3 of your program.
5. Insert the file name of the subroutine you want in line zero.
6. RUN the program to merge the subroutine.
7. When the merging stops, type CLOSE 1 and press RETURN.
WARNING:
Watch your line numbers! It is your responsibility to ensure that the line numbers of
the incoming subroutine do not conflict with those in your program. If there is a
conflict, the incoming line of the subroutine will replace the matching program line.
This little routine takes advantage of the auto-editing capabilities of your computer.
Line two reads the next program line from the diskette file and displays it at the top
of the screen. GET# rather than INPUT# is used to read the file to prevent colons and
commas from causing us to lose parts of lines. Line three prints the message GOTO 1
below the program line, homes the cursor, stuffs two carriage return characters into
the keyboard buffer and ends the program. The carriage returns in the buffer enter
the new program line and cause the execution of the GOTO 1 line. When execution
resumes on line one, the file has been closed in the computer but is still open in the
drive. The POKE in line one reopens the file in the computer. Line two gets and
displays the next line and so on. The last line in the ASCII file will be READY. This
halts execution with either an OUT OF DATA ERROR or SYNTAX ERROR
message. The merge is complete. Delete lines zero to three and you are all done.
Sequential Program Files
Mike Todd reported an interesting tidbit of information about Commodore computers
in his DISK FILE column in the May 1983 issue of the ICPUG Newsletter. He
reported that, because of the way Commodore computers interpret SAVE and LOAD
164
commands, it is possible to SAVE a program as a sequential file. Not only can you
SAVE it, you can even LOAD it again! Here are the commands to use:
SYNTAX:
SAVE "-filename, SEQ, W",device#
LOAD "f ilename,SEQ,R",device#
EXAMPLES:
SAVE "HIDDEN STUFF, SEQ, W" , 8
LOAD "HIDDEN STUFF, S,R" ,8
F*=" SUPER FORCE, S,R"
LOAD F*
WHERE:
filename = the file name (up to 16 characters); may be contained in a
string variable.
SEQ = file type, may be abbreviated with S.
W = WRITE mode used to save hidden program file.
R = READ mode used to load hidden program file.
device# = the disk drive device number.
When you SAVE a program in this way, it appears in the directory as a SEQ file.
However, this is not an ASCII text file. The program is in exactly the same form as it
would be in a normal PRG file. That's why it can be loaded again. This appears to be a
neat way of hiding programs on a diskette because everybody knows that you have to
use GET# or INPUTS to read a sequential file.
7.8 Typical Application
For each of the chapters on file handling (Chapters 6, 7, 8 and 9), I have developed a
mailing list program to illustrate a practical use for the techniques you have learned.
Only those portions of the program that deal with data handling are changed from
one chapter to the next. The portions relating to data entry, editing and printouts
remain the same.
In this chapter, the mailing list program uses a sequential file for storing the data
about the club members. The data is read into an array at the beginning of the
program. A revised file is written just before the program terminates.
To illustrate the greater flexibility you have when the data is all held in memory, the
program allows you to sort the list either alphabetically or by zip code. The Shell-
Metzner sort used is a considerably modified adaptation of the one given by Raeto
West in Programming- the PET/CBM, p.134 (Compute! Books).
Background information:
You've just been elected secretary of a local civic organization. As secretary, you are
responsible for making mailing labels for the monthly newsletter that goes out to all
500 members. You would like to use your new computer to take some of the drudgery
out of the task.
165
Analysis of the problem:
You need a simple mailing list program to keep track of the names and addresses of
the 500 members. The program should allow you to add new members, modify
members' names and addresses, produce a list of members and produce a set of
mailing labels.
Planning a record:
The first step in planning the program is to decide what information we need about
each member. You've checked over the club records and made up the following
summary.
Field
Description
Typical Case
Worst Case
1
2
3
4
5
6
Last name
First name
Street address
City
State
Zip code
8 characters
6 characters
18 characters
8 characters
2 characters
6 characters
14 characters
9 characters
23 characters
18 characters
2 characters
6 characters
Total
48 characters
72 characters
Record organization:
Each record will consist of the six fields indicated above. The record size will vary.
We will use carriage return characters as field delimiters in this application.
Let's go over the main parts of the program to see how it works:
Line Range Description
1000-1 1 10 Initialize variables and ask user to insert disk.
1130-1220 Open command channel and old data file. If there is no existing data
file, branch to the menu.
1240-1360 Read the number of records in the old file and then loop to read the
existing data into arrays. As data is read, check whether file is
unsorted, sorted alphabetically or sorted by zip code.
1400-1550 Display main menu, input option chosen and branch to the
appropriate part of the program.
1590-1680 Increment count of existing members, set flag to indicate an
unsorted file, and input the data for the new member.
1720-1910 Input the number of the record to edit, set the flag to indicate an
unsorted file, display the existing data and input any changes.
1950-2110 Print a list of members on the screen or printer.
166
2150-2470 Print mailing labels on the screen or printer. If the user presses Q,
the printing is suspended after the next label is printed and the user
has the option of aborting the print job or restarting it at any point.
This is helpful for paper jams!
2510-2650 Replace the existing data file with a new updated file, close the files
and end the program.
2700-2950 A Shell-Metzner sort of the mailing list data. Depending on the
value of FL, the list is sorted alphabetically or by zip code.
*****************
PROGRAM: MAIL SEQ
*****************
0 REM 1541 USER'S GUIDE SECTION 7.8
1 REM COPYRIGHT: G. NEUFELD, 1984
2 :
3 REM MAIL LIST PROGRAM USING SEQUENTIAL
FILE
4 :
lOOO REM* MAILING LIST PROGRAM (SEQ FILE
)
1005 DIM NL*<50>,NF*(50>,AD*<50>„CI*<50>
,SA*(50),ZP*<50>
lOlO PX*=CHR*<17):REM PRINTER UPPER/LOWE
R CASE
1020 FL=2:REM CURRENT SORT STATE
1030 SM*<0>= "SORTED BY ZIP CODE"
1040 SM*(l)="SORTED ALPHABETICALLY"
1 050 SM* ( 2 ) = " UNSORTED "
1060 PRINT" CCLR* MAILING LIST PROGRAM"
1070 PRINT" {DOWN} INSERT DATA DISKETTE IN
DISK DRIVE"
1080 PRINT" {DOWN* PRESS <RVS>RETURN{ROFF>
WHEN READY"
1090 :
llOO GET A*: IF A*<>"" GOTO HOO
1110 GET A*: IF A*OCHR*<13) GOTO lllO
1120 :
1130 REM OPEN COMMAND CHANNEL & INITIALI
ZE
1140 OPEN 15,8, 15, "10"
1150 INPUT#15,E,E*,T,S
1160 IF E>19 THEN PRINT E;E*;T;S:CL0SE15
:STOP
1170 :
1180 REM OPEN SEQUENTIAL FILE
1190 OPEN 1,8,7, "SEQ MAIL DATA,S,R"
1200 INPUT#15,E,E*,T,S
167
1210 IF E=62 THEN NR=0: CLOSE 1:G0T0 1400
:REM NO OLD FILE
1220 IF EM9 THEN PRINT E;E*;T;S:CL0SE15
:STOP
1230 :
1240 REM READ NUMBER OF RECORDS AT START
OF FILE
1250 INPUT#1,NR
1260 :
1270 REM READ EXISTING RECORDS INTO ARRA
YS
1280 FOR K=l TO NR
1290 : INPUT#1,NL*(K>,NF*<K>,AD*<K>,CI*<
K),SA*<K),ZP*<K>
1300 IF NL*<KXNL*(K-1)THEN AF=1
1310 IF ZP*(KXZP*(K-1)THEN ZF=1
1320 NEXT
1330 IF AF=1 AND ZF=1 THEN FL=2:REM UNSO
RTED
1340 IF ZF=0 THEN FL=0:REM SORTED BY ZIP
1350 IF AF=0 THEN FL=1:REM SORTED ALPHAB
ETICALLY
1360 CLOSE 1
1370 :
1380 REM* DISPLAY MENU
1390 :
1400 PRINT" CCLR>CRVS> CDOWN> MAILING LIST
MENU "NR"{LEFT} RECORDS"
1410 PRINT" {DOWNM. ENTER NEW MEMBER"
1420 PRINT" CD0WNJ2. EDIT EXISTING RECORD
II
1430 PRINT" {DOWN} 3. PRINT MEMBERS LIST"
1440 PRINT" < DOWN] 4. PRINT MAILING LABELS
fa
1450 PRINT" {DOWN} 5. SORT ALPHABETICALLY"
1460 PRINT" {DOWN} 6. SORT BY ZIP CODES"
1470 PRINT" {DOWN} 7. TERMINATE PROGRAM"
1480 PRINT" {DOWN} WHICH OPTION (1-7)"
1490 PRINT" {DOWN} <RVS> LIST IS "SM*<FL>"
II
1500 GET A*: IF A*=""GOTO 1500
1510 A=ASC(A*>-48
1520 IF A<1 OR A>7 GOTO 1500
1530 IF A=5 THEN FL=1
1540 IF A=6 THEN FL=0
1550 ON A GOTO 1590,1720,1950,2150,2700,
2700, 2480
1560 :
168
1570 REM: ENTER NEW MEMBER
1580 :
1590 NR=NR+l:REM ADD 1 MEMBER
1600 FL=2:REM NO LONGER SORTED
1610 PRINT" <CLRJ {DOWN} ENTER DATA FOR M
EMBER #"NR
1620 NL*<NR)="?": INPUT" tDOWN>LAST NAME: "
;NL$<NR)
1630 NF*(NR)="?": INPUT" £DOWN>FIRST NAME:
";NF*<NR>
1640 AD* <NR)="?": INPUT" {DOWN} STREET ADDR
ESS:";AD*<NR)
1650 CI* <NR>="?": INPUT" {DOWN}CITY/TOWN: "
;CI*(NR)
1660 SA* (NR) ="?" : INPUT" {DOWN}STATE: " ; SA*
(NR)
1670 ZP*(NR)="?": INPUT" {DOWN}ZIP CODE:";
ZP*(NR)
1680 GOTO 1400
1690 :
1700 REM* EDIT EXISTING MEMBER
1710 :
1720 PRINT" tCLRXDOWN} EDIT EXISTING MEM
BER{DOWN}"
1730 M*=" RECORD NUMBER (l-"+MID*<STR*(NR
>,2>+">"
1740 PRINTM*;" O"
1750 PRINT" {UP} "TAB <LEN<M*));: INPUT X*:X
=INT(VAL(X*))
1760 IF X<1 OR X>NR GOTO 1400
1770 :
1780 PRINT" {CLR> {DOWN} EXISTING DATA FOR
MEMBER #"X
1790 PRINT" CDOWNJ LAST NAME: ";NL*<X>
1800 PRINT" {DOWN}FIRST NAME: ";NF*(X>
1810 PRINT" {DOWN} STREET ADDRESS: ";AD*<
X)
1820 PRINT" {DOWN} CITY/TOWN: ";CI*(X>
1830 PRINT" {DOWN} STATE: ";SA*(X>
1840 PRINT" {DOWN} ZIP CODE: ";ZP*(X>
1850 PRINT" {HOME* (DOWN 3> "TAB < 10) : INPUT
NL*<X)
1860 PRINT"{D0WN}"TAB<11): INPUT NF*(X)
1870 PRINT "{DOWN} "TAB (15): INPUT AD*(X>
1880 PR I NT " {DOWN} " TAB (10): INPUT CI*(X)
1890 PRINT" {DOWN} "TAB <6): INPUT SA*<X)
1900 PRINT "{DOWN}" TAB < 9): INPUT ZP*(X)
1910 GOTO 1400
1920 :
1930 REM* PRINT LIST OF MEMBERS
169
1940 :
1950 PRINT" {CLR}{D0WN} PRINT LIST OF MEM
BERS"
1960 INPUT" {DOWN* OUTPUT TO SCREEN OR PR I
NTER (S/P) S CLEFT 3>";Z*
1970 PR I NT "{CLR} {DOWN} LIST OF MEMBERS"
1980 PRINT:DN=3:IF LEFT*<Z*,1)="P" THEN
DN=4
1990 OPEN 4,DN
2000 FOR K=l TO NR
2010 : IF DN=4 THEN PRINT#4,PX*;
2020 : PRINT#4, RIGHT* <" "+STR*<K> , 3) ; "
"NF*<K>" "NL*<K>
2030 : IF DN=4 THEN PRINT#4,PX*j
2040 : PRINT#4, " "AD*<K>", "CI*<K)%
"SA*(K)", "ZP*<K>
2050 NEXT
2060 CL0SE4
2070 IF DN=4 GOTO 1400
2080 PRINT" {DOWN} PRESS {RVS} RETURN {ROF
F} FOR MENU"
2090 GET A*: IF A*<>"" GOTO 2070
2100 GET A*: IF A*OCHR*<13> GOTO 2100
2110 GOTO 1400
2120 :
2130 REM* PRINT MAILING LABELS
2140 :
2150 PRINT" {CLR} {DOWN} PRINT MAILING LAB
ELS"
2160 INPUT" {DOWN} OUTPUT TO SCREEN OR PR I
NTER (S/P) SCLEFT 3}";Z*
2170 PRINT "{CLR} {DOWN} PRINTING MAILING
LABELS"
2180 PRINT" {DOWN} PRESS Q TO SUSPEND PRIN
TING {DOWN}"
2190 PRINT:DN=3:IF LEFT*<Z*, 1 )="p" THEN
DN=4
2200 OPEN 4,DN
2210 FOR K=l TO NR
2220 : IF DN=4 THEN PRINT#4,PX*?
2230 : PRINT#4,NF*(K>" "NL*<K)"„"
2240 : IF DN=4 THEN PRINT#4,PX«;
2250 : PRINT#4,AD*(K>","
2260 : IF DN=4 THEN PRINT#4,PX*;
2270 : PRINT#4,CI*<K>", "SA* (K) ", "ZP*(K
)
2280 : IF DN=4 THEN PRINT#4,PX«;
2290 : PRINT#4:PRINT#4
2300 : GET A*: IF A*="Q" GOTO 2390
2310 NEXT
2320 CL0SE4
170
2330 IF DN=4 GOTO 1400
2340 PRINT" {DOWN> PRESS <RVS> RETURN {ROF
F> FOR MENU"
2350 GET A*: IF A*<>"" GOTO 2350
2360 GET A*: IF A*OCHR*<13> GOTO 2360
2370 GOTO 1400
2380 :
2390 PRINT "i DOWN > OUTPUT ABORTED"
2400 PRINT" {DOWN* COMMANDS: A = ABORT"
24 iO PRINT" R = RESTART"
2420 GET A*: IF A*="A"THEN CL0SE4:60T0 14
OO
2430 IF A*<>"R" GOTO 2420
2440 PRINT" <DOWN> RECORD #"K"HAS JUST BEE
N PRINTED"
2450 PRINT" {DOWNiRESTART LIST AT RECORD
# "K+l
2460 PRINT" £UP> "TAB (24): INPUT K:PRINT
2470 GOTO 2230
2480 :
2490 REM* TERMINATE PROGRAM
2500 :
25 IO REM OPEN SEQUENTIAL FILE FOR REPLAC
E
2520 OPEN 1,8,7, "@0:SEQ MAIL DATA,S,W"
2530 :
2540 PRINT#1,NR:REM NUMBER OF RECORDS
2550 :
2560 FOR K=l TO NR
2570 : PRINT#1,NL*<K)
2580 : PRINT#1,NF*<K)
2590 : PRINT#i„AD*(K>
2600 : PRINT#1,CI*<K>
2610 : PRINT#1,SA*<K>
2620 : PRINT#1,ZP*<K)
2630 NEXT K
2640 :
:END
ALPHABETIC
2650 CLOSE 1:
2660 :
2670 REM SHELL-MET ZNER SORT
ALLY
2680 REM FL=1 ALPHABETIC FL=0 BY ZIP CO
DE
2690 :
2700 PRINT" {CLRMDOWN 2>S0RTIN6. . . PLEAS
E BE PATIENT"
2710 M=NR
2720 M=INT(M/2)
IF M=0 GOTO 1400: REM SORT COMPLETED
2730
2740 J=l:K=NR-M
171
2750 I=J
2760 L=I+M
2770 :
27SO IF FL=0 SOTO 2840
2790 :
2800 IF NL*<IXNL*<L> GOTO 2940
2810 IF NL*<I)=NL*<L>ANDNF*<IX=NF*(L> G
OTO 2940
2820 GOTO 2860
2830 :
2840 IF ZP*(IX=ZP*(L> GOTO 2940
2850 :
2860 H*=NL* ( I ) : NL* < I ) =NL* (L) : NL* <L) =H*
2870 H*=NF* < I ) : NF* < I ) =NF* (L) : NF* <L> =H*
2880 H*=AD* ( I ) : AD* ( I ) =AD* (L) : AD* (L> =H*
2890 H*=CI* ( I ) : CI* ( I ) =CI* (L) : CI* <L> =H*
2900 H*=SA* ( I ) : SA* ( I ) =SA* (L) : SA* (L) =H*
2910 H*=ZP* ( I ) : ZP* ( I ) =ZP* <L) : ZP* <L> =H*
2920 1=1 -M
2930 IF I>0 GOTO 2760
2940 J=J+l:IF J>K GOTO 2720
2950 GOTO 2750
Possible improvements and modifications:
1. Provide retrieval of a record for editing by name rather than by the record
number which changes when the file is resorted.
2. Provide selective printouts of lists or labels based on the contents of certain fields.
For example, only those members who have zip codes between 60004 and 60120.
Or, if your members live in Canada, those members with postal codes between
S4K 3N2 and S6K 1L6.
172
Chapter
8
RELATIVE FILES
Despite what you may have read in some popular magazines about problems with
relative files on Commodore single disk drives, relative files are reliable. However,
you must always specify the byte parameter in the position command and make sure
that the disk drive has time to position to the record properly if you are writing or
modifying records in a random sequence.
In this chapter we take a first look at relative files and their use. Relative files are
somewhat more complex than program or sequential files but learning to use them is
worth it. They are fast and quite reliable.
Chapter 9 takes a look at some more advanced applications of relative files. Section
13.13 describes how these files are stored on a diskette and explains just what those
mysterious side sectors are all about.
8.1 Introduction to Relative Files
A relative file is a special type of data file. It is somewhat more complicated to use and
understand than a sequential data file. However, it is a very useful file type because
you can read or write any record in the file without disturbing the rest of the file.
Relative files are particularly useful for storing information that needs to be changed
or updated frequently.
One limitation of a relative file is that all the records must be the same length. When
you create a relative file, you must specify the record length. All the records in the file
will be this length whether or not all the space is needed. If the record length is large
and the actual record is small, diskette storage space is wasted. If you make the
record length small, you will inevitably end up with a record that won't fit. You must
plan carefully to make sure that your longest record will fit. Many programmers feel
that the ability to read and write individual records in the file more than compensates
for this limitation.
When to use a relative file:
1. If your data must be revised or updated frequently.
2. You have more data than you can read into memory at once.
173
Unless your application meets both of these criteria, you would probably be better off
using a sequential file.
The benefits of using relative files are:
1. Individual records within a file can be read or written.
2. Access to a record is fast provided you know the record number.
The drawbacks of using relative files are:
1. Storage space is wasted because the record length is fixed.
2. Usually you have to maintain one or more sequential files that hold a list of the
record numbers in the order needed to access the records in alphabetical order or
some other order. These are known as index files. Chapter 9 deals with indexed
relative files.
Some typical uses of relative files are:
1. Mailing or membership lists.
2. Inventory control.
3. General data base management.
8.2 Structure of a Relative File
A relative file is composed of many records. Each record usually stores all the infor-
mation about a single person, place or thing. The number of characters in each record
is the same. This is called the record length and must be specified when the file is first
created. A typical file might look like this:
Record #1
Record #2
Record #3
Record #4
x>
Since each record occupies the same amount of space, there may be a considerable
amount of wasted space, particularly if the amount of actual data varies a great deal
from record to record. However, the fixed record length makes it easy for the Disk
Operating System (DOS) to determine exactly where any given record starts and
stops. As a user of relative files you don't have to concern yourself about how this is
done using side sectors and other pointers. All you have to do is specify the number of
the record that you want to read or write and the disk drive takes care of all the messy
details. If you are interested in the messy details, read through Section 13.13.
8.3 Planning a Record
Whenever you want to use a relative file in a program, you should spend some time
preplanning the record length and layout. There are two general approaches to
record layouts, fixed length fields and variable length fields.
Fixed length fields:
When a record is organized using fixed length fields, each field is always a particular
174
length and occupies the same position in a record. For example, suppose we used
fixed length fields to store a last name, a first name, a street address, a city, a state
and a zip code in a record for a mailing list program. We might decide that our
record would be laid out this way:
ield
Field Contents
Field Length
Characters
#1
Last name + delimiter
14 characters
1 to 15
#2
First name + delimiter
9 characters
16 to 25
#3
Address + delimiter
22 characters
26 to 48
#4
City/town + delimiter
19 characters
49 to 68
#5
State/Prov.+ delimiter
2 characters
69 to 71
#6
Zip code + delimiter
6 characters
72 to 78
Diagrammatically our record would look like this:
Field 1 Field 2 Field 3 Field 4 Field 5 Field 6
1 IS 25 48 68 71 78
1 1 1 i—l 1-
Jones *Tim *23 First St. *Anytown ♦AS»123456*
1 1 1 1 i (-
* = carriage return character CHR$(13)
Variable length fields:
A second and usually better way to organize the record would be to use variable
length fields. With this method the fields are not padded with blanks to make them
any particular length. Only a field delimiter separates the fields. By using this
approach, we can usually get away with a shorter record length. This is because the
person with the longest last name usually does not have the longest first name and
address. In the case of our mailing list we might well decide to use a record length of
66 (60 characters + 5 field delimiters + 1 record delimiter).
Diagrammatically our record would look like this:
Fields vary in length as needed ■< Nulls at end
Jones*Tifl»23 First St.*Anytown*AS»123456*
* = carriage return character CHR$(13)
One limitation of this approach is that you have to be careful not to overfill the record.
Always design your program so it checks the total length of the actual record before
you write it.
When to use fixed length fields:
1. If the number of characters of actual data for each field does not vary much. A
good example would be social security numbers that are always the same length.
175
2. If you need to read only selected portions of complex records. By using fixed
length fields you can position within the record to the start of the field you want to
read.
3. When convenience and speed of accessing parts of records is more important than
conserving diskette storage space.
When to use variable length fields:
1. If the number of characters of actual data for each field varies .considerably. A
good example would be names or addresses in a mailing list.
2. When conserving diskette storage space is more important than convenience or
speed of accessing parts of records.
Remember that any commas or carriage return characters you are using as separators
(delimiters) between fields or records also count as characters in the record.
8.4 Deciding on the Record Length
Unfortunately, there is no magic formula for determining the optimum record length.
You just have to make an intelligent guess after studying the data you want to store.
If you make the record length large, you will never have to worry about overfilling a
record but you will waste a great deal of diskette storage space. On the other hand, if
you make the record length small, you may end up with some records that won't fit.
If you are using fixed length fields, you will have to determine the length of each field
independently. In general, each field must be large enough to hold the worst case data
for that field. For example, if the last names of the members of your organization
vary in length from five (Smith) to 14 (Anadranastacki) characters, the last name
field will have to be at least 15 characters long (remember that the delimiter counts).
Once you have determined the worst case lengths for each field, add up the field
lengths to get the record length.
If you are using variable length fields, you will have to make an intelligent guess. The
optimum field length is somewhere between the sum of the typical field lengths and
the sum of the worst case field lengths. Often the average of these two sums is a
good guess.
WARNING:
When planning the record length for your file, you should be aware that there are
three forbidden record lengths. These forbidden lengths are 42, 58 and 63 characters
(see Section 8.5).
8.5 Using a Relative File
Using a relative file is quite easy. Just follow these steps:
1. Open the command channel.
2. Open the relative file.
176
3. Use the P command to position to the record you want.
4. Write information into the record using PRINT#.
5. Use INPUT# or GET# to read information from the record.
6. Close the relative file and the command channel when you are finished using
the file.
Let's look at each of these steps in a bit more detail.
Opening the Command Channel
You must open the command channel so that you can specify the record you want to
access using the P (position) command. You will also need the command channel open
to check for disk errors (see Section 5.10 about how to do this). The statement that
opens the command channel looks like this:
SYNTAX :
OPEN file#,device#,cmd chnl#
ALTERNATE:
OPEN -f i le#, device** ,cmd chnl#, "command"
EXAMPLES:
OPEN 15,8,15
OPEN 15,8, 15, "10"
CN=7 : DN=8 : CC= 1 5 : C*= " SO : OLD F I LE "
OPEN CN,DN,CC,C*
WHERE:
file# = the logical file number (any integer 1-127; often 15 is used
as the file#).
device# = the device number (normally eight),
cmd chnl# = 15; the command channel number,
command = any disk command.
NOTE:
All examples use file# 15 for the command channel. The file# for the command
channel may be any integer from one to 127. However, 15 is used in all examples
because this makes it easier to remember which file# to use for disk commands.
Opening the Relative File
An OPEN command is used to tell the disk drive the name of the file you wish to use.
On the Commodore 64 or the VIC-20, the OPEN command to access a relative file
looks like this:
SYNTAX:
OPEN -file#, device*, channel**, "-filename, L, "
+CHR*(size>
(may be used for either new or existing -files)
177
alternate:
OPEN file#,device#,channel#, "filename"
(may be used -for existing files only)
examples:
OPEN 5,8,4," CUSTOMERS , L , " +CHR* < 20 )
OPEN 1,8,5," BUNCHASTUFF , L , " CHR* ( 1 25 )
OPEN 5,8,3,"0LD D*"
F*= " MEMBER DATA " : RS=85
OPEN 2,8,12,F*+",L,"+CHR*<RS>
WHERE:
file# = the logical file number (any integer 1-127).
device# = the device number (normally eight).
channel# = the channel number (any integer 2-14).
filename = the file name (up to 16 characters); may contain wild
cards if file already exists.
L = identifies that a record length follows; optional if the file
already exists,
size = the number of characters in a record. May be any integer
2-254 except 42, 58 or 63; optional if the file already exists.
WARNINGS:
1. Certain record lengths are forbidden. Although you are not warned about this in
the 1541 User's Manual, there are three forbidden record lengths. These forbidden
record lengths are 42, 58 and 63.
2. An OPEN statement that specifies any of these forbidden lengths will be rejected
by the DOS with a SYNTAX ERROR. The reason is that these lengths are the
ASCII values of characters that have special meaning to the DOS (42 ="*", 58 = ":",
and 63 = "?").
CAUTION:
The record length must be specified when a new relative file is created. Once you
have created the file, its record length cannot be changed. If you try to specify a new
record length for an existing file, your request will be ignored and the file will not be
opened. The only way to change the record length is to open a new relative file using
the new record size and copy the information into the new file one record at a time!
NOTES:
1. You do not have to specify whether you want to read or write to a relative file.
Relative files are always opened in READ/WRITE mode.
2. Be very careful about punctuation. The comma that follows the L must be there if
you are specifying the record length. If you are not going to specify the record
length, close the quotes after the file name like this: "filename".
178
HINT:
If you are opening a new relative file, you should create the entire file at once. You do
this by using the POSITION command to position to the highest record number you
expect to have in the file. This forces the disk drive to create an entire file of blank
records. You do not have to do this. However, it is good practice to do this because the
file will be as compact as possible. This makes finding records faster and reduces the
wear and tear on your drive.
REMINDERS:
1. Monitor the error status of the disk drive. You should use the command channel to
monitor the error status of the disk drive. Be sure to check the status whenever
you open a file to make sure that the file has been opened properly. If the file is not
open, your data will be lost! See Section 5.10 about how to do this.
2. All open files must have different logical file numbers. Since the logical file
number is used to identify one particular file, you must use different logical file
numbers for each of the files that are open simultaneously. Once you close a file,
you are free to reuse that logical file number in your program.
3. All open files must have different channel numbers. Since the channel number
(secondary address) identifies a particular data communication channel in the
disk drive, each file that is open must have a unique channel number. Once you
close a file, you are free to reuse that channel number in your program.
4. You can only have a few files open at once. When you are using a relative file,
you are very limited in the number of other disk files you can have open. The
limits are:
1 Relative + 1 Sequential, or
1 Relative + 1 Sequential + 1 Direct-Access, or
1 Relative + 2 Direct- Access
ANNOTATED EXAMPLES:
OPEN 1,8,5," BUNCH AD AT A , L , " +CHR* ( 20 )
Opens file# 1 to disk using channel* 5 to access or create a new relative file named
BUNCHADATA. Each record is 20 characters long.
OPEN 3,8,9,"PILEAJUNK"
Opens file# 3 to disk using channel* 9 to access an existing file named LOTASTUFF.
If this is a relative file, you will be in READ/WRITE mode. The record length will be
the size specified in the OPEN statement used to create the file.
OPEN 7,8,12," LOTASTUFF , L , " +CHR* ( 66 )
Opens file# 7 to disk using channel* 12 to access or create a new relative file named
LOTASTUFF. Each record is 66 characters long.
179
Positioning to a Record
Since you can read or write any record in the relative file, you must specify the
number of the record you want. You do this using a P (position) command. Because
the VIC-20 and the Commodore 64 have BASIC 2.0, you have to specify the record
number in lo byte/hi byte form. You can calculate the lo byte and hi byte like this:
Hi byte = INT(Record Number/256)
Lo byte = Record Number -256 * Hi byte
NOTES:
1. INT stands for integer. The INT function on Commodore computers always rounds
a number down to the next lower whole number. Here are a few examples of how
it works:
INT(6.236) = 6 INT(0.3695) = 0 INT(53.589) = 53
INT(1.995) = 1 INT(-2.814) = -3 INT(-0.012) = -1
2. Be careful of the order of operations. The multiplication operation is performed
before the addition operation.
Correct order Incorrect order
LO = 675 -256 * 2 LO = 675 -256 * 2
LO = 675 -(256 * 2) LO =(675 -256)* 2
LO = 675 -512 LO = 419 * 2
LO = 163 LO = 838
EXAMPLES:
Record number = 110
Hi byte = INT(110/256)
= INT(0.4296875)
= 0
Lobyte = 110-0*256
= 110-0
= 110
Record number = 725
Hi byte = INT(725/256)
= INT(2.83203125)
= 2
Lobyte = 725-2*256
= 725-512
= 213
180
Once you have calculated the lo and hi bytes of the record number, you use them in
the P (position) command. On the Commodore 64 or the VIC-20 the command looks
like this:
syntax:
PRINT#-Fi 1 e#, "P" ; CHR* (channel #) 5 CHR* <r 1 ) ; CHR* (rh) ;
CHR* (byte)
EXAMPLES:
PRINT#15, "P" ; CHR* <5) ; CHR* (7) ; CHR* <0) ; CHR* ( 1 )
PRINT#15, "P" ; CHR* (9) ; CHR* <5> ; CHR* < 1 > ; CHR* <20)
RH= INK RN/ 256 ) : RL=RN-256*RH : B Y=24
PRINT#15, "P" ; CHR* <3) ; CHR* <RL) ; CHR* <RH) ; CHR* <BY)
WHERE:
file# = the file number of the command channel.
channel* = the relative file's channel number.
rl = the lo byte of the record number.
rh = the hi byte of the record number.
byte = the position within the record where you want the read or
write operation to begin. Byte may be any integer between
one and the record length.
WARNINGS:
1. The byte position must be specified! The byte parameter must always be present
in a POSITION statement. It is not optional as indicated on page 35 of the 1541
User's Manual If this parameter is omitted, the carriage return character,
CHR$(13), that is sent to the drive is sometimes assumed to be the position par-
ameter and, if it is, you will be positioned to start reading or writing at position 13
in the record. Beware!
2. In some cases positioning may be inaccurate! When you position to a record in a
relative file to write or rewrite the record, the position command may not work
properly. This only -occurs when you access files randomly and you write or rewrite
the record. For more details and a fix see Section 8.6.
NOTE:
The P command is a direct command to the disk so it is issued over the command
channel. As a result, you must have opened the command channel (OPEN 15,8,15)
previously.
ANNOTATED EXAMPLES:
PRINT#15, ,,P";CHR*(7);CHR*<2);CHR*(0>;CHR*(1)
In the relative file opened using channel* 7, position to record number 2 (2 + 0*256),
byte number one.
181
RN=623
RH=INT(RN/256):REM HI BYTE OF RECORD NUMBER
RL=RN-256*RH :REM LO BYTE OF RECORD NUMBER
PRINT#15, "P" 5 CHR* (3) 5 CHR* (RL) ; CHR* (RH) ; CHR* ( 1 )
In the relative file opened using channel* 3, position to record number 623 (111 +
2*256), byte number one.
RN=1286
RH=INT(RN/256):REM HI BYTE OF RECORD NUMBER
RL=RN-256*RH :REM LO BYTE OF RECORD NUMBER
PRINT#15, "P" ; CHR* (9) ; CHR* (RL) J CHR* <RH) 5 CHR* (27)
In the relative file opened using channel* 9, position to record number 1286 (6 +
5*256), byte number 27.
Writing a Record
Once you have positioned to the record you want, use a PRINT* statement to write
the record. It is a good idea to write out the entire record using a single PRINT#
statement (see WARNING below). You assemble all the parts of the record into one
long string and then print out the string like this:
REM OPEN COMMAND CHANNEL AND FILE
OPEN 15,8,15
OPEN 7,8,12," LOT ASTUFF , L , " CHR* ( 66 )
■
REM POSITION TO THE START (BYTE=1) OF RECORD #K
RH=INT(K/256):REM HI BYTE OF RECORD NUMBER
RL=K-256*RH :REM LO BYTE OF RECORD NUMBER
PRINT#15, "P" ; CHR* ( 12) ; CHR* (RL) ; CHR* (RH) ; CHR* ( 1 )
■
REM ASSEMBLE RECORD
C*=CHR*(13):REM FIELD SEPARATOR
S*=NAME* (K) +C*+ADDR* (K) +C*+CITY* (K) +C*+ZIP* (K)
REM WRITE OUT RECORD
PRINT#7,S*
WARNING:
The PRINT* statement always writes to a record from your current position in the
record to the end of the record! This means that you cannot write out just the first
part of a record and leave the last part untouched. Here is an example to show you
what happens:
Suppose we wanted to change Charles' phone number.
182
Sm
ith
Charles
453-1248 Oct. 15/47
1
5
10 15
22 25 30 35 40
Original
Byte position
Position to byte #22 and PRINTS the new number 789-1238.
Fin*11 Smith Charles 789-1238 ///nulls///
Byte position i 5 10 15 22 25 30 35 40
The end of our record has been wiped out with nulls!
CAUTION:
Be sure to check the length of your data before you write it to the file. If your data is
longer than the record length, the data will be written into the file but will be
truncated to the proper size to protect the following record. The drive will report a 51,
OVERFLOW IN RECORD, error message. Remember that delimiters, including
the final one at the end of the record, count as characters in the record.
NOTES:
1. The P command is a direct command to the disk so it is issued over the command
channel. As a result, you must have opened the command channel (OPEN 15,8,15)
previously.
2. You do not have to pad your data with blanks to make it the same length as the
record length. The disk operating system will automatically fill the rest of the
record with null characters, CHR$(0)'s. This can cause problems (see WARNING
above).
Reading a Relative File
Whenever you open a relative file you are automatically positioned to the start of the
first record in the file. The P command is used to position to the record you want. As
noted above you must always specify the byte position within the record where you
want to start reading. Once you have positioned to the record you want, you may use
either an INPUTS or GET# statement to read information from the record. In general,
plan to use INPUTS rather than GETS because INPUTS reads more information at
one time. However, there are some situations when you must use GETS to examine a
file closely.
When to Use GETS:
1. To check the contents of a file that won't read properly.
2. To read a file that contains stray commas, colons or null bytes that cause problems
for INPUTS.
Whether you use INPUTS or GETS you always start reading at the position you
specified in the P statement. As you read fields or individual characters from the
record, the disk drive will keep track of where you are in the record. You can only
183
read forward within a record. You cannot back up. If you have read past a piece of
data that you need, you must reposition with another P command.
Normally data is read from a relative file as it is needed. Since any record may be
accessed quickly, there is no need to read the data into an array.
Most programmers use an INPUT# statement since you get more information at
once. If you used carriage return characters as field separators, you may use either a
single INPUTS statement or a series of INPUT# statements, one for each field. If you
used commas as field separators, you must use a single INPUT# statement to input
all the information at once. Here's how to read the record we wrote earlier.
100 REM OPEN COMMAND CHANNEL AND FILE
110 OPEN 15,8,15
120 OPEN 7,8,4,"L0TASTUFF,L,"+CHR*<66>
130 REM POSITION TO START OF RECORD #K
140 RH=INT<K/256):REM HI BYTE OF RECORD
NUMBER
150 RL=K-256*RH :REM LO BYTE OF RECORD
NUMBER
160 PRINT#15, "P" ; CHR* <4) ; CHR* <RL) ; CHR* (R
H);CHR*<1>
170 :
180 REM INPUT ALL INFORMATION AT ONCE METHOD #1
190 INPUT#7, NAME* (K) , ADDR* (K) , CITY* <K> , Z
IP*(K)
180 REM INPUT FIELDS SEPARATELY METHOD #2
185 REM SEPARATOR IS CHR* (13)
190 INPUT#7,NAME*(K)
200 INPUT#7,ADDR*(K)
210 INPUT#7,CITY*(K)
220 INPUT#7,ZIP*<K>
WARNING:
You can read past the end of a record. When you attempt to write a record that is
longer than the record size, you are warned (51, OVERFLOW IN RECORD) and the
record is truncated to protect the next record. However, there is no similar mechanism
to prevent you from reading past the end of a record. The only way to tell that you
have reached the end of a record is by checking the status variable, ST (see NOTES
below).
NOTES:
1. The position command works fine for reading a record. As long as you specify the
byte parameter in the POSITION command, you will always be positioned to the
correct record and be able to read it correctly.
2. You can use ST to check for the end of a record. Whether you use GET# or
INPUTS, you may use the status variable, ST, to check for the last non-null
184
character in a record (see Section 5.15). When you have reached the end of the
record, ST will have the value 64.
REMINDERS:
1. If your file contains a CHR$(0), it will not be read correctly. The value stored into
the string variable is not CHR$(0). The null string (no characters) is stored there
instead. This is the reason for the common GET# statement:
GET#1,X*ZIF X*="" THEN X*=CHR*(0>
2. INPUT# always reads to a carriage return or null character. Since an INPUT#
statement will always read as far as the next carriage return or null character,
you must be careful not to skip over values and lose information (see Section 5.14).
3. Using GET# can cause garbage collection problems. If you use GET# for reading
a relative file you have to build up long strings using concatenation (X$=X$+A$).
This can cause problems on a VIC-20 or Commodore 64 because you store a lot of
unused fragments of strings in memory. These slowly fill up RAM memory. When
the memory is full, the computer has to sort through the mess and discard all the
unused strings. This is called garbage collection. It takes from five seconds to 20
minutes to do. While garbage collection is taking place, the cursor goes away and
nothing seems to happen.
Closing the File
Once you have finished using the relative file, you must close it. If you forget, you may
lose data from the last few records you wrote and the BAM may be incorrect! A close
statement looks like this:
SYNTAX:
xx CLOSE file#
EXAMPLES:
75 CLOSE 1
600 CLOSE 5: CLOSE 15
WHERE:
xx = the line number.
f ile# = the logical file number you used when you opened the file.
CAUTION:
Be sure to close your relative file. Closing the file is important when you have been
writing to your relative file. If you forget to close the file, the data block that contains
the last record you wrote may not be recorded onto the diskette. However, failing to
close a relative file is less disastrous than failing to close a program or sequential file.
Because you are in read/write mode rather than write mode, you cannot create an
unclosed relative file (*REL).
185
NOTES:
Although the following procedure does not appear to be necessary when closing a
relative file on the 1541, following it guarantees that you will not have the problem of
the last record never being recorded on the diskette.
1. Never CLOSE a relative file that is less than two full sectors (508 bytes) long. If
necessary, create several empty records by positioning to the start of record number
l+(508/record size).
2. Just before closing the file, position to the start of the first record. Unless you have
violated rule #1, this will force a write to disk and your file is safe.
8.6 The Problem with Relative Files and How to Fix It
Several articles have been published recently reporting problems with relative files
on the 1541 and 2031 Commodore single disk drives. Various fixes were proposed.
Since I had not encountered any problems when using relative files, I was interested
in trying to duplicate the findings and possibly find a better fix. Unfortunately, most
of the articles did not define the conditions under which the problems occurred
(record size, record number, reading or writing or updating a record, etc.), nor did
they provide listings of programs that produced errors. The articles just spoke in
general terms about "problems with the POSITION command." This made it very
difficult to localize the problem.
I began by writing a few simple test programs and found that the byte parameter
was definitely not optional when using relative files on the 1541. Maybe that was the
problem. I developed a comprehensive test program to check this out. The program
systematically worked through all possible record sizes. For each record size it would
create a test file. The number of records in the file was determined by the record size.
As the record size increased, the number of records decreased so that each file
contained about 20,000 characters (bytes) of data. The file was created by writing the
records sequentially (record #1, #2,...#N). The records were designed so that you could
identify where you were in the file by reading three consecutive bytes. The bytes were
the record size (RS), the low byte of the record number (RN), and the byte position
within the record (BY). For example, record #3 in a file with a record size of 12 would
look like this:
RS RN BY RS RN BY RS RN BY RS RN BY
12 3 3 12 3 6 12 3 9 12 3 12
$0C $03 $03 $0C $03 $06 $0C $03 $09 $0C $03 $0C
Once the file was written, 100 record numbers and byte positions were chosen at
random. These records were accessed and read. If a positioning error occurred, a
detailed report was printed. Once the program was written and checked, the test
began. After 36 hours of steady running, the tests were completed.
The results? One very hot 1541 that was in desperate need of an alignment. My
printout indicated the three forbidden record lengths. However, not a single
positioning error was found. I was ready to conclude that any positioning problems
186
were the result of programmers forgetting to specify the byte parameter in the
POSITION command or alignment problems with their drives.
However, I decided to do some more testing and developed the program listed at the
end of this section. This program makes it easy to do a wide variety of tests on relative
files. You can create a new file of records, revise all the records, or read and check all
the records. You can even choose the order in which the records are written, revised
or read. They can be done normally (1, 2,..., N), in reverse order (N, N-l,..., 1), or in
scrambled order (25, 2,...). Using this program I was finally able to generate the kind
of errors described in the articles. I was also able to define the exact conditions under
which problems do occur and try various fixes. Here are the results:
When do positioning problems occur?
The only time the POSITION command does not work with complete reliability is
when you are writing or revising records in a scrambled order.
How frequently do errors occur?
The error rate is not consistent. Approximately one percent of the records will not be
revised at all and another one percent will be revised incorrectly. Small files seem to
have a higher error rate than large files.
Is there a pattern of errors?
The errors do not fall into an easily discernable pattern. However, a record that spans
two blocks on a diskette rarely causes problems.
What seems to be the problem?
The Disk Operating System (DOS) uses two data buffers to hold records as they are
being read from or written to the diskette. When a record that spans two sectors is
being read or written, one buffer holds the start of the record and the other buffer
holds the last part of the record. However, when records that are contained entirely
within one block are being read or written, the DOS alternates between buffers (one
active and the other inactive). It appears that during write operations the file manager
side of the DOS may begin printing new information into the buffer before the disk
controller side of DOS has finished reading in the sector that holds the appropriate
record.
What is the fix?
By introducing a short time delay between the POSITION command and the PRINTS
command the error rate can be reduced considerably. However, to ensure complete
reliability you should issue the POSITION command twice and then delay for a short
time before executing the PRINTS command that sends the revised data to the disk
drive. The lines that you need are:
480 REM CONVERT RECORD # TO LO/HI FORM
490 RH=INT(N<K)/256) :RL=N(K>-256*RH
187
500 :
510 REM POSITION ONCE
520 PRINT#15, "P"CHR*<5>CHR*<RL>CHR*<RH>C
HR* < 1 )
530 :
540 REM POSITION A SECOND TIME
550 PRINT#15, "P"CHR*<5>CHR*<RL)CHR*<RH)C
HR* < 1 )
560 :
570 REM PAUSE (30 JIFFIES = .5 SECONDS)
580 T2=TI+30
590 IF TKT2 GOTO 590
600 :
610 rem print to the file
620 print#i,m*;cr*;n<k>
The two POSITION commands followed by the time delay seem to give the floppy
disk controller side of the DOS enough time to read in the desired record from the
diskette. With the test program the delay slows down the updating of the file by about
30 percent. With the delay it took about 15 minutes and 25 seconds for the computer
to revise 600 records in scrambled order (1.54 sec/record) and there were no errors.
Without the delay the same process took only 11 minutes and 25 seconds (1.14
sec/record), but there were three errors. In most practical applications the 0.4 second
delay per record should be virtually undetectable because most of the time the com-
puter is waiting for the operator to type in the revisions. The operator rarely waits for
the computer.
Here is a listing of the test program that I used for checking the operation of relative
files. You may want to use it to do a bit more testing on your own. Maybe you can
come up with a better, and simpler, fix.
*******************
program: rel tester
*******************
This program is designed to make the creation, revision and checking of test files
easy. You may want to try various alternatives for the fix described. Just substitute
your favorite fix for lines 780-810. To change the number of records or the record size,
simply change lines 110 and 120.
0 REM 1541 USER'S GUIDE SECTION 8.6
1 REM COPYRIGHT: 6- NEUFELD, 1984
2 :
3 REM CREATE AND TEST RELATIVE FILES
4 :
lOO REM REL TESTER
110 NR=30:REM NUMBER OF RECORDS
120 RS=25 :REM RECORD SIZE
130 M*<l)="ORIGINAL #"
140 M*(2)="MODIFIED #"
188
ISO CR*=CHR*<13>
160 DIM N(NR>
170 :
180 PRINT" {CLR}REL FILE TESTER"
190 PRINT" {DOWN} INSERT TEST DISKETTE"
200 PRINT" {DOWN}OPTIONS: "
210 PRINT" {DOWN} 1. CREATE FILE"
220 PRINT" 2. MODIFY FILE"
230 PRINT" 3. READ FILE"
240 INPUT "{DOWN} YOUR CHOICE (1-3) 2{LEF
T 3}";X*
250 X=ASC<X*>-48
260 IF X<1 OR X>3 GOTO 180
270 :
280 OPEN 15,8,15, "10"
290 INPUT#15,E,E*,T,S
300 IF E>19 THEN PRINTE;E$; T;S:STOP
310 IF X=l THEN PRINT#15, "S: SHORT REL":P
RINT#15, "10"
320 :
330 PRINT" {DOWN} NORMAL, REVERSE, OR SCRA
MBLED ORDER"
340 INPUT" {DOWN} <N/R/S> SO-EFT 3}";0*
350 IF 0*0"N" AND 0*0"S" AND 0*<>"R" 6
OTO 330
360 :
370 IF 0*="R"THEN F0RL=1 TO NR:N(L)=NR-L
+l: NEXT: GOTO 490
380 :
390 PRINT" {DOWN} MAKING LIST OF RECORD NU
MBERS"
400 F0RL=1 TO NR:N(L)=L:NEXT
410 IF 0*="N" GOTO 490
420 :
430 PRINT" RANDOMIZING LIST"
440 FOR R=NR TO 1 STEP- l: REM SCRAMBLE SE
QUENCE
450 : rx=1+int<rndc1)*r>
460 : h=n<r):n<r)=n<rx>:n(RX>=h
470 NEXT
480 :
490 OPEN 1,8,5, "SHORT REL,L, "+CHR*(RS>
500 INPUT#15,E,E*,T,S: PRINT" {DOWN}REL FI
LE OPEN -"E;E*
510 IF E> 19 THEN PRINT"FATAL ERROR" :STO
P
520 IF X=3 GOTO 890: REM TO READ FILE
530 :
540 IF X=2 GOTO 670
550 :
189
560 INPUT" CDOWN>POSITION TO END (Y/N) Y
CLEFT 3>"5W*
570 W=ASC(W*):IF W=78 GOTO 670
580 IF W<>89 GOTO 560
590 :
600 REM POSITION TO END TO CREATE ENTIRE
FILE
610 RH=INT(NR/256):RL=NR-256*RH
620 PRINT#15, "P"CHR*(5)CHR*(RL)CHR*(RH)C
HR* ( 1 )
630 INPUT#15.E,E*
640 PRINT"POSITION TO END -"E;E*
650 PRINT#1,"DUMMY RECORD"
660 :
670 INPUT" CDOWN>READ BEFORE WRITE (Y/N)
Y CLEFT 3>";Y*
680 IF Y*<>"Y" AND Y*<>"N" GOTO 670
690 :
700 REM WRITE OR MODIFY RELATIVE RECORDS
710 PRINT" {DOWN} CONTENTS = "M*(X)
720 PRINT"CUP}"TABUO>;: INPUT M*
730 FOR K=l TO NR: PRINT M*;N(K)
740 RH=INT (N (K) /256> : RL=N <K) -256*RH
750 PRINT#15, "P"CHR*(5)CHR*(RL)CHR*(RH)C
HR* ( 1 )
760 IF Y*<>"Y" GOTO 820
770 :
780 REM POSITION AGAIN AND DELAY
790 PRINT#15, "P"CHR*(5)CHR*(RL)CHR*(RH)C
HR* < 1 )
800 T2=TI
810 IF TKT2+30 GOTO 810
820 PRINT#1,M*;CR*;N(K)
830 NEXT
840 INPUT"CDOWN}READ BACK (Y/N) YCLEFT
3> " ; R*
850 IF R*="Y" THEN PR I NT: GOTO 890
860 CLOSE l: CLOSE 15: END
870 :
880 REM READ RELATIVE RECORDS
890 FOR K=l TO NR
900 RH= I NT ( N ( K ) / 256 ) : RL=N ( K ) -256*RH
910 PRINT#15, "P"CHR*(5)CHR*(RL)CHR*(RH)C
HR* ( 1 )
920 TH=INT (N (K) *RS/254) : Z*=STR* (TH+1 )
930 IF THOINT((l+N(K))*RS/254)THENZ*=Z*
+" SPAN"
940 TNPUT#1,A*,B*:PRINT"#"N(K)"IS: ";A«;
B*;TAB(30);Z*
190
950 IF T*=A* AND MID* (STR* <N<K> > ,2>+" "=
B* GOTO lOOO
960 :
970 PRINT"* BAD * PRESS RETURN" :NE=NE+i
980 GET R*:IF R*OCHR*<13> GOTO 980
990 PRINT" CUP} CUP}"
lOOO NEXT
1010 CLOSE 1: CLOSE 15
1020 PR I NT "THERE WERE "NE" ERRORS"
REMINDER:
The fix for the POSITION command is only needed when writing records in a
scrambled order. If you are reading the file or writing records in sequential order,
the fix is not required.
8.7 Replacing an Existing Relative File
Normally you would never want to replace an entire existing relative file with a
revised version. Instead, you would modify only those records that need changing.
The only way to eliminate an existing relative file is to scratch the entire file. Once the
old file is scratched, you are free to open a replacement file like this:
250 OPEN 15,8,15:REM
OPEN COMMAND CHANNEL
260 PR I NT# 1 5 , " SO : DATA " : REM
SCRATCH OLD FILE
270 INPUT#15,E,E*,T,S:REM
READ ERROR CHANNEL
280 IF E<20 GOTO 310: REM
CHECK FOR DOS ERROR
290 PRINT E;E*;T;S:REM
PRINT ERROR MESSAGE
300 CLOSE 15: STOP: REM
ABORT ON ERROR
310 OPEN 3,8,5, "DATA, L,"+CHR* (50): REM
OPEN FILE
320 REM PROGRAM CONTINUES...
NOTE:
Replace does not work with relative files! The technique of replacing the file by
putting @0: in front of the file name in the OPEN statement does not work for a
relative file. If you try it, the @0: will be ignored and the file opened for normal
read/write access. You must SCRATCH a relative file.
191
8.8 Adding Data to a Relative File
Whenever you open a relative file you are free to position to any existing record in the
file and read or write it. However, when you are adding information to the file, you
may be positioning to records that don't exist yet. When you do this the 1541 disk
operating system will report a 50, RECORD NOT PRESENT, error message.
Surprisingly, you will not get this error for each new record you add. You only get
this error when a new block (sector) must be added to the file. If you are adding new
data to the file, this error is normal and can be ignored. It merely indicates that the
file is being expanded to include the new record(s).
The only time to be concerned about a 50, RECORD NOT PRESENT, error message
is if that record should really exist already. In this case, you should check for bugs in
your program. If you get this error when trying to access the last few records in your
file, you probably forgot to close the file the last time you used it and have lost the last
few records.
CAUTION:
Before you append more data to a file, check to be sure that you have enough free
space on the diskette to hold the longer file. If you run out of space, the last few
records will not be written. However, with a relative file you do not lose your old data
when this happens. See Section 5.13 for a method of checking the number of blocks
free from within a program.
REMINDER:
It is good practice to create the entire file at once. You can improve the speed of
accessing a relative file slightly if the data is stored on one area of the disk rather than
being scattered around the disk surface. When it is stored in one area the read/write
head doesn't have to move around as much. It is simple to reserve a block of space on
disk for your file when you first create it. Simply position to the highest record
number you expect to have in the file. This forces the disk drive to create all the
records at one time, from #1 to the number you specified. Your file will be as compact
as possible. As a bonus, you will never have to worry about whether your file is at
least two blocks long (see NOTES about closing a relative file in Section 8.5), or about
50, RECORD NOT PRESENT, error messages.
8.9 Typical Application
For each of the chapters on file handling (Chapters 6, 7, 8 and 9), I have developed a
mailing list program to illustrate a practical use for the techniques you have learned.
Only those portions of the program that deal with data handling are changed from
one chapter to the next. The portions relating to data entry, editing and printouts
remain the same.
In this chapter, the mailing list program uses a relative file for storing the data about
the club members. A record is read when the data is needed. Only the record currently
being added, edited or printed is held in memory. This reduces the demands on RAM
memory.
192
Records are stored in the relative file in the order in which they are entered. There is
no provision for sorting the list either alphabetically or by zip code. To retrieve a
record for editing, you must know its record number. The record numbers are given
on the membership list printout.
Background information:
You've just been elected secretary of a local civic organization. As secretary, you are
responsible for making up mailing labels for the monthly newsletter that goes out to
all 500 members. You would like to use your new computer to take some of the
drudgery out of the task.
Analysis of the problem:
You need a simple mailing list program to keep track of the names and addresses of
the 500 members. The program should allow you to add new members, modify
members' names and addresses, produce a list of members, and produce a set of
mailing labels.
Planning a record:
The first step in planning the program is to decide what information we need about
each member. You've checked over the club records and made up the following
summary.
Field
1
2
3
4
5
6
Description
Typical Case
Worst Case
Last name
8 characters
14 characters
First name
6 characters
9 characters
Street address
18 characters
23 characters
City
8 characters
18 characters
State
2 characters
2 characters
Zip code
6 characters
6 characters
Total
48 characters
72 characters
Record organization:
Each record will consist of the six fields indicated above. We will use variable length
fields. This allows us to use a somewhat shorter record. We anticipate that our longest
record will contain about 60 characters (the average of 48 and 72). As a result, we will
use a record size of 66 (60 characters + 5 field delimiters + 1 record delimiter). A
typical record will look like this:
10
— ►—
20
30
40
h-
50
60
66
Jone&*Tim«23 First St.*Anytown«AS*123456*////padded with nulls////
' = carriage return character
One limitation of using variable length fields is that we have to be careful not to
overfill the record. The program must always check the total length of the actual
record before we write it, to make sure we don't lose the end of the record.
193
Let's go over the main parts of the program to see how it works:
Note the fix is used only when records are being edited. It is not needed for initially
writing the records (done sequentially) or reading the records.
Line Range Description
1000-1070 Initialize variables and ask user to insert disk.
1100-1170 Open command channel and data file.
1200-1250 Position to record #1 and input the number of records in the file.
1290-1390 Display main menu, input option chosen and branch to the appropriate
part of the program.
1430-1650 Increment count of existing members, input the data for the new
member, assemble the record and check it for length. Position to
record #1 to update the record count and then store new record in
its place.
1690-2110 Input the number of the record to edit, position to the record and read
it. Display the existing data, input any changes and rewrite the record.
Note the fix when rewriting the record in lines 2000-2060.
2150-2330 Print a list of members on the screen or printer.
2370-2690 Print mailing labels on the screen or printer. If the user presses Q,
the printing is suspended after the next label is printed and the user
has the option of aborting the print job or restarting it at any point.
This is helpful for paper jams!
2730-2790 Close the file and end the program.
*****************
PROGRAM: MAIL REL
*****************
0 REM 1541 USER'S GUIDE SECTION 8.9
1 REM COPYRIGHT: G. NEUFELD, 1984
2 :
3 REM MAIL-LIST PROGRAM USING RELATIVE F
ILE
4 :
lOOO REM* MAILING LIST (RELATIVE FILE)
1010 R*=CHR*(13):REM CARRIAGE RETURN CHA
RACTER
1020 PRINT" {CLR> MAILING LIST PROGRAM (R
EL) "
1030 PRINT" {DOWN* INSERT DATA DISKETTE IN
194
DISK DRIVE"
1040 PRINT" {DOWN} PRESS CRVS J RETURN {ROFF>
WHEN READY"
1050 :
106O GET A*: IF A*<>"" GOTO 1060
1070 GET A*: IF A*OCHR*(13> GOTO 1070
10GO :
1090 REM OPEN COMMAND CHANNEL St INITIAL I
ZE
llOO OPEN 15,8, 15, "IO-
ll lO INPUT#15,E,E*,T,S
1120 IF E>19 THEN PRINT E;E4; T; S: CL0SE15
:STOP
1130 :
1140 REM OPEN RELATIVE FILE
1150 OPEN 1,8,7, "MAIL DATA,L, "+CHR«<66>
1160 INPUT#15,E,E*,T,S
1170 IF E>19 THEN PRINT E;E«;T;S:CLOSE15
: STOP
1180 :
1190 REM POSITION TO RECORD #1
1200 PRINT#15, "P"+CHR* (7) +CHR* ( 1 ) +CHR* (O
> +CHR* ( 1 )
1210 INPUT#15,E,E*,T,S
1220 :
1230 REM INPUT NUMBER OF EXISTING RECORD
S
1240 INPUT#1,A*:IF A*=""THEN A*="0"
1250 NR=VAL(A$>
1260 :
1270 REM* DISPLAY MENU
1280 :
1290 PRINT" {CLR>CDOWN> MAILING LIST MENU
"NR" RECORDS"
1300 PRINT" {DOWNM. ENTER NEW MEMBER"
1310 PRINT" {DOWN* 2. EDIT EXISTING RECORD
1320 PRINT" (DOWN* 3. PRINT MEMBERS LIST"
1330 PRINT" {DOWN} 4. PRINT MAILING LABELS
1340 PRINT" {DOWN} 5. TERMINATE PROGRAM"
1350 PRINT" CDOWN>WHICH OPTION (1-5)"
1360 GET A*: IF A*=""GOTO 1360
1370 A=ASC<A*)-48
1380 IF A<1 OR A>5 GOTO 1360
1390 ON A GOTO 1430,1710,2150,2370,2700
1400 :
1410 REM: ENTER NEW MEMBER
1420 :
1430 NR=NR+l:REM ADD 1 MEMBER
195
1440 PRINT" <CLR>{DOWN> ENTER DATA FOR M
EMBER *"NR
1450 NL*="?": INPUT" < DOWN} LAST NAME:";NL«
1 460 NF*= '"?"Z INPUT " CDOWN> F I RST NAME : " ; NF
*
1470 AD*="?": INPUT" CDOWN> STREET ADDRESS:
" 5 AD*
1480 CI*="?": INPUT" {DOWNJCITY/TOWN: ";CI*
1490 SA*="?" : INPUT" €DOWN>STATE: " ; SA*
1500 ZP*="?": INPUT" {DOWN>Z IP CODE:";ZP*
1 5 1 O F*=NL*+R*+NF*+R*+AD*+R*+C I *+R*+SA*+
R*+ZP*
1520 IF LEN<F*><60 GOTO 1560
1530 PRINT" <DOWN>ERROR: RECORD IS TOO LO
NG":X=NR
1540 FOR K=l TO 5000: NEXT: GOTO 2130
1550 :
1560 REM POSITION TO RECORD #1
1570 PRINT#15, "P"+CHR* <7) +CHR* ( 1 ) +CHR* (O
) +CHR* ( 1 )
1580 PRINT#1,NR
1590 :
1600 REM POSITION TO RECORD # NR+1
1610 NH = INT(<NR+l)/256>
1620 NL - <NR+1) - 256*NH
1 630 PR I NT# 1 5 , " P " +CHR* ( 7 ) +CHR* < NL ) +CHR* (
NH)+CHR*<1)
1640 PRINT#1,F*
1650 INPUT#15,E,E*,T,S
1660 IF E>19 AND E050 THEN PRINT E;E«?T
;S: CLOSE 15: STOP
1670 GOTO 1290
1680 :
1690 REM* EDIT EXISTING MEMBER
1700 :
1710 PRINT" <CLR}<DOWN> EDIT EXISTING MEM
BER<DOWN>"
1720 M*=" RECORD NUMBER (l-"+MID* (STR* (NR
>,2>+">"
1730 PRINTM*; " O"
1740 PRINT" CUP> "TAB <LEN<M*> >;: INPUT X*:X
=INT(VAL(X*>)
1750 IF X<1 OR X>NR GOTO 1290
#X
1760
m
1770
REM POSITION TO REC
1780
m
1790
NH = INT<<X+l)/256)
1800
NL = (X+l) - 256*NH
196
1810 PRINT#15, "PM+CHR*<7)+CHR*<NL)+CHR*<
NH)+CHR*<1)
1820 INPUT#1 , NL*, NF*, AD*, CI*, SA*. ZP*
1830 PRINT" <CLR>£DOWN> EXISTING DATA FOR
MEMBER #"X
1840 PRINT" {DOWNJLAST NAME: ";NL*
1850 PRINT" <DOWN>FIRST NAME: ";NF*
I860 PRINT" {DOWN} STREET ADDRESS: ";AD*
1870 PRINT" {DOWN}CITY/TOWN: ";CI*
1880 PRINT" {DOWN* STATE: ";SA*
1890 PRINT" {DOWN} ZIP CODE: ";ZP*
1900 PRINT" {HOME} {DOWN 3} "TAB (10) : INPUT
NL*
1910 PRINT "{DOWN}" TAB(ll): INPUT NF*
1920 PRINT" {DOWN} "TAB (15): INPUT AD*
1930 PRINT" {DOWN} "TAB (10): INPUT CI*
1940 PRINT" {DOWN} "TAB <6): INPUT SA*
1950 PR I NT "{DOWN}" TAB (9): INPUT ZP*
1960 REM POSITION TO RECORD #X
1970 NH = INT((X+l)/256)
1980 NL = (X+l) - 256*NH
1990 :
2000 REM FIX FOR POSITION PROBLEM
2010 :
2020 PR I NT# 1 5 , " P " +CHR* < 7 ) +CHR* < NL ) +CHR* (
NH)+CHR*<1)
2030 PRINT#15, "P"+CHR*(7)+CHR*<NL)+CHR* (
NH)+CHR*(1)
2040 :
2050 T2 = TI + 30
2060 IF TKT2 GOTO 2060
2070 :
2080 F*=NL*+R*+NF*+R*+ AD*+R*+C I *+R*+SA*+
R*+ZP*
2090 IF LEN(F*)>59 GOTO 1530
2100 PRINT#1,F*
21 lO GOTO 1290
2120 :
2130 REM* PRINT LIST OF MEMBERS
2140 :
2150 PRINT" <CLR> {DOWN> PRINT LIST OF MEM
BERS"
2160 INPUT" {DOWN} OUTPUT TO SCREEN OR PRI
NTER <S/P) SO.EFT 3}";Z*
2170 PRINT:DN=3:IF LEFT*(Z*, 1)="P" THEN
DN=4
2180 OPEN 4,DN
2190 FOR K=l TO NR
2200 REM POSITION TO RECORD #X
2210 NH = INT((K+l)/256)
197
2220 NL = (K+l) - 256*NH
2230 PR I NT# 1 5 , " P " +CHR* ( 7 ) +CHR* ( NL ) +CHR* <
NH)+CHR*<1>
2240 INPUT#1 , NL*, NF*, AD*, CI*, SA*, ZP*
2250 PRINT#4,RIGHT*(" "+STR* <K> , 3) ; " "
NF*" "NL*
2260 PRINT#4," "AD*", "CI*", "SA*", "
ZP*
2270 NEXT
2280 CL0SE4
2290 IF DN=4 GOTO 1290
2300 PRINT" CDOWN> PRESS €RVS> RETURN {ROF
F> FOR MENU"
23 lO GET A*: IF A*<>"" GOTO 2290
2320 GET A*: IF A*OCHR*(13> GOTO 2320
2330 GOTO 1290
2340 :
2350 REM* PRINT MAILING LABELS
2360 :
2370 PRINT" {CLR} {DOWN} PRINT MAILING LAB
ELS"
2380 INPUT" {DOWNIOUTPUT TO SCREEN OR PRI
NTER (S/P> SCLEFT 3>"jZ*
2390 PRINT" €DOWN>PRESS Q TO SUSPEND PRIN
TING"
2400 PRINT: DN-3: IF LEFT* (Z*, 1 >="P" THEN
DN=4
2410 OPEN 4,DN
2420 FOR K=l TO NR
2430 REM POSITION TO RECORD #X
2440 NH = INT(<K+l>/256)
2450 NL = CK+1) - 256*NH
2460 PR I NT# 1 5 , " P " +CHR* < 7 ) +CHR* < NL ) +CHR* (
NH)+CHR*(1>
2470 INPUT#1,NL*,NF*,AD*,CI*,SA*, ZP*
2480 PRINT#4,NF*" "NL*","
2490 PRINT#4,AD*","
2500 PRINT#4,CI*", "SA*", "ZP*
2510 PRINT#4:PRINT#4
2520 GET A*: IF A*="Q" GOTO 2610
2530 NEXT
2540 CLOSE4
2550 IF DN=4 GOTO 1290
2560 PRINT" {DOWN} PRESS {RVS} RETURN {ROF
F} FOR MENU"
2570 GET A*: IF A*<>"" GOTO 2570
2580 GET A*: IF A*OCHR*(13) GOTO 2580
2590 GOTO 1290
2600 :
2610 PRINT" {DOWN} OUTPUT ABORTED"
198
2620 PRINT" (DOWN} COMMANDS: A = ABORT"
2630 PRINT" R = RESTART"
2640 GET A*: IF A*="A"THEN CLOSE4:60TO 12
90
2650 IF A*<>"R" GOTO 2640
2660 PRINT" { DOWN > RECORD #"K"HAS JUST BEE
N PRINTED"
2670 PRINT" {DOWN} REST ART LIST AT RECORD
# "K+l
2680 PRINT" <UP> "TAB (24): INPUT K: PRINT
2690 GOTO 2440
2700 :
27 lO REM* TERMINATE PROGRAM
2720 :
2730 REM POSITION TO RECORD #20
2740 NH = O
2750 NL = 20
2760 PRINT#15, "P"+CHR* (7) +CHR* (NL) +CHR* (
NH) +CHR* ( 1 )
2770 INPUT#15,E,E*,T,S
2780 IF E>19 AND E<>50 THEN PRINT E;E*;T
;s: close is: stop
2790 CLOSE l: CLOSE 15
2800 :
Possible modifications and improvements:
1. Provide a means of recalling records by member's name or zip code rather than
by record number.
2. Produce membership lists and mailing lists that are sorted alphabetically or by
zip code.
3. Provide a way of deleting a member from the list. At present, the only way to
delete a member is to call up his or her record for editing and enter the data for a
new member.
199
CHAPTER
9
INDEXED RELATIVE FILES
NOTE:
This chapter is not designed for beginners. The reader is assumed to be reasonably
familiar with both sequential and relative files. The sections dealing with the binary
chop should be useful for anyone who has to search ordered lists or arrays.
This chapter takes a more advanced approach to file handling. One or more sequential
files are used as indices into a large data base held in a relative file. Many of the
commercial data base management systems make use of the techniques discussed in
this chapter.
9.1 Introduction to Indexed Relative Files
An indexed relative file is not a special type of data file. The term refers to a file-
handling technique that uses a relative file to store data and one or more sequential
files to store indices to the main relative file. This is one of the most powerful file-
handling techniques available to micro users.
Programming for an application using indexed relative files is somewhat more com-
plex than using either sequential or relative files by themselves. You must maintain a
large, relative file that contains the data and one or more sequential files that index
the information in the relative file in several ways. Unless these indices are carefully
maintained, you will have a mess! However, careful attention to detail will pay off.
This technique allows you to store large quantities of data and read or update it in a
variety of ways quickly and easily. Some ambitious programmers have built systems
that allow you to have several diskettes full of data. A few programmers have even
produced programs that allow for variable length records in the main relative file.
However, multi-diskette data banks and creating variable length records in a relative
file are beyond the scope of this book, and are recommended only for those hackers
with masochistic tendencies!
When to use an indexed relative file:
1. If you have very large amounts of data (over 30K).
2. The data must be revised or updated frequently.
3. You must be able to access the data in a variety of ways.
201
The benefits of using indexed relative files are:
1. They can provide very fast read/write access to any record.
2. They can handle very large amounts of data.
3. They can generate reports in a variety of ways such as by listing members either
alphabetically or by zip codes.
The main drawback of using indexed relative files is the complexity of the program-
ming required. Very careful attention to detail is required to program an application
using this approach. It is very easy to ruin an index by failing to update it after some
operation such as adding or editing a record. Extensive testing is required before you
can be confident your program will work correctly.
Some typical uses of indexed relative files are:
1. Mailing or membership lists.
2. Bibliographies.
3. School or medical records.
4. General data base management.
9.2 Organization of the Files
The data is stored in a large, relative file composed of many records. One record
usually stores all the information about a single person, place or thing. The records in
the file are not sorted. They are simply entered one after another. A data file composed
of last names and telephone numbers might look like this:
Record #1 Record #2 Record #3 Record #4
3
Smith ,7281495*
Jones, 4532659*
Pauls, 4531875*
Adams, 4533514*
* = carriage return character
In this application there would probably be two sequential files that would serve as
indices to the relative file. One file would contain the record numbers in the sequence
required to list the records in alphabetic order by name. The second file would
contain the record numbers in the sequence required to list the records in numeric
order by telephone number. The two sequential, index files for our very simple four
record relative file would look like this:
Alphabetical order file Numerical order file
4*2*3*1*
3*2*4*1 •
* = carriage return character
If we wanted to generate a list of the records in alphabetic order, we would read and
print out the records from the relative file in the order specified in the alphabetical
order file. If we wanted the list in numeric order, we would read and print out the
records from the relative file in the order specified in the numerical order file.
202
9.3 Planning an Application of Indexed Relative Files
Whenever you want to use an indexed relative file in a program, you should spend
some time preplanning both the main, relative file and the index files. It is much
easier to fix mistakes before you begin programming than to patch your program
later.
The Main Relative File
Since this is a normal relative file, you will have to consider record length and layout.
You may use either fixed or variable length fields. In most cases, variable length
fields are preferred because of the additional flexibility they provide (the worst case
for a last name may really be 16 characters instead of the 14 you counted on). Be sure
to make your records long enough!
If you have more than 254 characters of information about each individual, place or
thing, it will not all fit into one record. In these situations you will have to decide
whether to use two consecutive records in one file or two separate files. If you choose
to use two consecutive records in a single file, you will have to program carefully.
However, this approach keeps all the information together so that you don't have to
access separate files to read all the information about one person, place or thing. If
you choose the two file approach, you will have to close one to use the other, because on
the 1541 you can have only one relative file open at a time.
The Sequential Index Files
The most important aspect of planning the index files is to decide how many different
ways you want to be able to access the data in the main, relative file. In other words,
how many different fields will be key fields and be indexed? The more different
indices you have, the more ways you can list or scan your data. However, each
additional index makes your program more complex. In general, plan on using at
least two different indices.
Normally, you don't have to limit the number of index files because of space limitations
on the diskette or in memory. These files are quite small, all you are storing in them is
a list of numbers. If for some strange reason you do think you might run short of
diskette space, you might want to convert the record numbers into lo byte/hi byte
form and store CHR$(lo) and CHR$(hi) in the sequential file. This allows you to index
up to 65,535 records using only two bytes for each. However, you will have to use
GET# to retrieve these bytes from the sequential file.
The best tactic for handling the indices is to read them all into arrays in memory at
the beginning of your program. This way they are always accessible for fast searching,
sorting or scanning. Just before your program terminates, you should rewrite these
index files so that you are always sure that they reflect any changes you have made in
the relative file. You may want to include a flag that indicates whether or not any
changes have been made and eliminate the rewriting of these files if no changes have
been made. However, this is not usually necessary because the index files are small,
and rewriting them does not take a long time.
An alternative approach would be to store both the record number and part or all of
the key field in the sequential index file. With the key field in RAM memory you can
search them without having to access the records on diskette. A memory search is
203
faster and there won't be as much wear and tear on your disk drive. However, this
technique uses more space both on the diskette and in RAM memory- As a result, this
technique is not appropriate for very large data bases.
9.4 Keeping an Index
A significant portion of any program that uses indexed relative files is dedicated to
updating the indices whenever a new record is added or a key field of an existing
record is modified. To get a feeling for the problems involved, let's consider adding a
new record to our four record name and phone number file.
The record we want to add is Brock,4532893.
Here are our existing files. The relative file will be on diskette while the two indices
will be in RAM.
Record #1
Record #2
Record #3
Record #4
Smith, 72B1495»
Jones, 4532659*
Pauls, 4531875*
Adams, 4533514*
Alphabetical order file Numerical order file
4*2*3*1*
3*2*4*1*
Our program will contain a record counter so it is easy to tell where to store the new
record in the main relative file. It is stored as record #5 (4+1) like this:
Record #1 Record #2 Record #3 Record #4 Record #6
Smith,72B1495* Jones, 4532659* Pauls,4531B75* Adams, 4533514* Brock, 4532S93*
Storing the record in the relative file was easy. The tough part is updating the
indices. How do we determine where Brock fits in among the other records? All we
have in memory is a list of record numbers. Obviously, we will have to read some of
the records in the relative file to find out.
The brute force method of determining Brock's position in the alphabetical order
index would be to start at the beginning of the index and read each corresponding
record and compare Brock with the name in the relative file. The process would go
something like this:
• Position to record #4 (first in the alphabetic list).
• Read name from record #4 (Adams).
o Compare "Adams" and "Brock" (not far enough yet).
» Position to record #2 (second in the alphabetic list).
• Read name from record #2 (Jones).
e Compare "Jones" and "Brock" (Eureka! we're there).
204
Shift all numbers in the list up one position, from position #2 on.
Original list: 4 * 2 * 3 * 1 *
After shift: 4*-* 2*3*1*
Insert new record number into position #2 in the list.
Modified list: 4 * 5 * 2 * 3 * 1 *
Now that we have modified the alphabetic list, we have to repeat the process for the
numeric list like this:
• Position to record #3 (first in the numeric list).
• Read phone number from record #3 (4531875).
• Compare "4531875" and "4532893" (not far enough yet).
• Position to record #2 (second in the numeric list).
• Read phone number from record #2 (4532659).
• Compare "4532659" and "4532893" (not far enough yet).
• Position to record #4 (third in the numeric list).
• Read phone number from record #4 (4533514).
• Compare "4533514" and "4532893" (Eureka! we're there).
Shift all numbers in the list up one position, from position #3 on.
Original list: 3*2*4*1*
After shift: 3*2*-* 4*1*
Insert new record number into position #3 in the list.
Modified list: 3 * 2 * 5 * 4 * 1 *
This same process would be followed each time you add a new record. The main
problem with this brute force method is that it is slow! The reason for this is the
number of records we have to read from the relative file. Accessing the diskette is
slow because we have to rely on the movement of mechanical parts. We could speed
up our program a great deal if we could reduce the number of diskette accesses we
need. The solution to our problem is the BINARY CHOP.
9.5 The Binary Chop
The binary chop is a very rapid way to scan a list. You may be familiar with it from
kids' guessing games. Have you ever played the game where one person thinks of a
number between one and 100, and the other person has to guess what it is? The way to
guess the number in the minimum number of tries is to reduce the number of
possibilities in half with each guess. Here's how it works with a number between one
and 64.
205
Player 1 Player 2
Think of a number from 1 to 64. Got one!
1 Is it larger than 32? (range=32) No!
2 Is it larger than 16? (range=16) Yes!
3 Is it larger than 24? (range=8) No!
4 Is it larger than 20? (range=4) No!
5 Is it larger than 18? (range=2) Yes!
6 Is it larger than 19? (range=l) Yes!
It must be 20! Yes!
By using the binary chop technique as we did in this game, you can always locate an
item in an ordered list very rapidly. The powers of two tell you the maximum number
of guesses you'll need:
List length: 64 128 256 512 1024 2048 4096 8192
Max guesses: 6 7 8 9 10 11 12 13
As you can see, the maximum number of guesses needed increases very slowly in
comparison to the size of the list. It is one of the most efficient ways of scanning a
large, ordered list
Let's apply the binary chop to scanning our relative file records. Consider the alph-
abetic list of record numbers. Here is our list
Alphabetic list: 4 * 2 * 3 * 1 *
We first check to see if our new record fits someplace between the smallest and the
largest values. In this case, record #4 is the smallest and record #1 is the largest. We
have to check these end points because our new record might fit before the first
existing record or after the last existing record.
•
Look up first index number in alphabetic list (it's four).
• Position to record #4 (smallest).
• Read name from record #4 (Adams).
• Compare "Adams" and "Brock" (doesn't come before the first record).
•
•
Look up last index number in alphabetic list (it's one).
Position to record #1 (largest).
Read name from record #1 (Smith).
Compare "Smith" and "Brock" (doesn't come after the last record).
Since our new record fits somewhere between the first and the last, we have deter-
mined upper and lower limits on its position:
lower limit = position #1 in list = smallest record = record #4
upper limit = position #4 in list = largest record = record #1
To apply the binary chop method, we then split the range in half by finding the
midpoint (average) between the positions of the records that are our upper and lower
limits.
Average of positions of upper and lower limit
= (l + 4)/2
= 2.5
= 2
In this case our average is not an integer. Since we can't position to record #2.5, we
had better drop the fraction and settle on two. We would use the integer function
(INT) to do this for us in a program.
• Look up the record# in position #2 of our alphabetic list (it's two).
• Position to record #2 (first average).
• Read name from record #2 (Jones).
• Compare "Jones" and "Brock" (too high).
The position of our new upper limit is two (since we were too high).
The position of our old lower limit is one (only changes if too low).
Split the range in half again by finding the midpoint (average) of the positions of our
new upper and lower limits.
Average of positions of upper and lower limit
= (l + 2)/2
= 1.5
= 1
At this point our average of the positions and the position of our lower limit are equal.
This happens only when the positions of our upper and lower limits are one apart.
This tells us that we have reached the end of our testing. The new value fits between
our upper and lower limits!
We have found where the new value goes, so insert it into the alphabetic list.
Shift all numbers in the list up one, from position #2 on.
Original list: 4*2*3*1*
After shift: 4*-* 2*3*1*
Insert new record number into the list.
Modified list: 4 * 5 * 2 * 3 * 1 *
In this situation, we didn't save any disk accesses by using the binary chop method.
However, we used a very small relative file. As the file gets larger and larger, the
binary chop begins to shine. Just take a look at this table:
Average Number of Disk Accesses to Scan an Ordered List
Length of List
Binary Chop*
Brute Force**
• Assumes two end point accesses plus others.
** Average number of disk accesses.
207
8
16
32
64
128
256
512
1024 2048 4096
5
6
7
8
9
10
11
12 13 14
4
8
16
32
64
128
256
512 1024 2048
The time and effort of programming the binary chop as illustrated in the typical
application at the end of this chapter is well worth it.
9.6 Programming the Binary Chop
The binary chop makes a very useful subroutine that you can insert into any program
in which you need to scan an ordered list. When you want to use it, you just pass it the
search image you want, and the routine will find where it should fit in the list.
Definition of variables used: .
SI$ = Search image (the item to be inserted).
RN = Record number (to be inserted into the index list).
IX() = Index list (one dimensional array).
NR = Number of items in old list (the starting upper limit).
FI$ = Image found in the file when we looked.
MAX = Current upper limit.
MIN = Current lower limit.
AVG = Average of lower and upper limit.
R = Record number to look up on disk (used by subroutine at
line 6000).
5000
3010
5020
5030
5040
5050
5060
5070
5080
5090
5100
5110
5120
5130
5140
5150
5160
5170
5180
5190
5200
5210
5220
5230
5240
5250
R=IXU>
SOSUB 6000
IF SI*<FI* THEN MAX=1 SOTO 5210
■
RntX(NR)
SOSUB 6000
IF SI*>FI* THEN MAX=NR+1:S0T0 5240
MIN-1 : MAX=NR
AVS = INT((MAX+MIN)/2)
IF AVS - MIN SOTO 5210
R=IX(AV6)
SOSUB 6000
IF SI*<FI* THEN MAX»AV6:G0T0 5110
IF SI*>FI* THEN MIN=AVG:GOTO 5110
MIN=AVG: MAX=AVS+1
FOR K=NR TO MAX STEP -1
: IX(K+1) - IX(K)
NEXT
IX (MAX) = RN
RETURN
6000 H=INT(R/256)
6010 L=R-RH*256
6020 PRINTttlS, "P"CHR*(4)CHR»(L)CHR* (H)CH
R*(l)
6030 INPUT#1,FI*,B*,C*,C*
6040 RETURN
Check if new fits before first
Find record # offirst in list
Read first record.
New fits before first
Check if new fits after last
Find record # of last in list
Read last record.
New fits after last.
We now know new fits somewhere
between first and last
Set lower & upper limits.
Find average of upper & lower.
We are done so jump out.
Find record # of average.
Read average record.
New fits before average so set upper
limit to average.
New fits after average so set lower
limit to average.
New equals file value so we put new
just after old.
Shift index numbers up one.
Insert new's record number.
Find hi byte of record #.
Find lo byte of record #.
Position to record.
Read record.
208
9.7 Alternate Uses for the Binary Chop
The binary chop is very versatile. Here are a few ways you might use it in different
situations:
1. Searching for one particular item in the file.
This is a very common application. It would be useful to scan the file to locate the
record# of an employee with a particular name, social security number or phone
number. Let's see how we can modify our listing to adapt it to this application. In line
5150 we check if the search image, SI$, is less than the image from the file, FI$. In
line 5170 we check if SI$ is greater than FI$. This means that in order for line 5180 to
be executed, SI$ must equal FI$. To convert our listing into a search for a particular
record, all we do is add a new line 5180 something like this:
5180 PRINT "RECORD #,,R"IS THE ONE YOU WANT": RETURN
If the condition in 5120 is satisfied, AVG=MIN, we know that the item we were
looking for is not in the file. To report this we could change line 5210 to read:
5210 PRINT "RECORD NOT FOUND IN THE FILE": RETURN
For this application, lines 5220 to 5250 are not required and should be deleted from
the listing.
2. Searching for a range of values in the file.
This is another common application. For example, you might want to find the record
numbers of all members having zip codes in the range of 60203 to 61258 or names
between JONES and SMITH. For this application we really have two situations, one
where we are looking for the top of our range, and the other for the bottom of our
range. Rather than try to make our subroutine handle both jobs at once, we will use
two calls to it and use a flag variable, FL, to indicate whether we want the upper limit
(FL=1) or the lower limit (FL=0).
The main modification to our routine has to do with what happens when the file
image, FI$, equals our search image, SI$. In this application, we may have several
records that produce the match. For example, if we are looking for all those individuals
having zip codes between 60000 and 61999, we will probably have several individuals
in zip code area 60000 and several in 61999. This means we can't just quit when we
find the first person with a particular zip code. We have to define the exact upper and
lower boundaries. Here's how we do it.
Change lines 5180 and 5190 to read:
5180 IF FL=0 THEN MAX=AVG:GOTO 5110 Looking for lower.
5190 IF FL=1 THEN MIN=AV6:G0T0 5110 Looking for upper.
These two lines force the routine to find the exact lower limit when FL=0 and the
exact upper limit when FL=1. Let's see how this works with a very simple example.
This is a small part of a file with 48 records. Our search image, SI$, is 60000 and
FL=0 (find lower limit).
209
#
FI$
Pass
MAX
MIN
AVG
FI$
SI$
Result
31
60201
1
48
1
24
59589
< 60000
MIN = AVG
30
60000
2
48
24
36
?
> 60000
MAX=AVG
29
60000
3
36
24
30
60000
60000
MAX=AVG
28
60000
4
30
24
27
60000
60000
MAX=AVG
27
60000
5
27
24
25
59789
< 60000
MIN = AVG
26
60000
6
27
25
26
60000
60000
MAX=AVG
25
59789
7
26
25
25
Quit since
MIN=AVG
24
59589
When our routine quits after looking for the lower limit (FL=0), MAX contains the
number of the first record that we want. If we were looking for the upper limit
(FL=1), MIN would contain the last record that we want. We can simplify our sub-
routine call slightly if we always return the boundary value in the same variable by
adding these lines.
5210 IF FL=0 THEN LIMIT - MAX
5220 IF FL=1 THEN LIMIT = MIN
5230 RETURN
For this application lines 5240 and 5250 are not required and should be deleted from
the listing.
9.8 Programming Considerations for indexed Relative Files
There are several things that you should consider when programming an application
that uses indexed relative files.
1. How large will the data base become?
Most data bases grow rapidly over time. If there is any possibility of your data base
growing so large that it won't all fit on a single diskette, you should plan your
program accordingly. One possible way to accomodate a large data base would be to
plan to store the program and the index files on diskette #1, and the relative file(s) on
diskettes #2, #3, etc. If you plan to use this approach, you must design your program
so that all files are closed except when data is actually being read or written. This will
allow you to swap diskettes. Be sure to remember to include the diskette number as
well as the record number in the indices.
2. How many different indices will I need?
It is always difficult to foresee all the different ways you might want to access or scan
your data. As a general rule, if you think that you might ever want your data indexed
in a particular way, include it from the start! However, if you find that you do need to
index your existing data base by a new field, you can do it. To create the new index
file you will need to go through the entire relative file record by record reading the
field of interest. The field followed by the record number is put into an array, the
array is sorted, and then the new index file is written out like this:
100 DIM N1KK) Dimension temp array.
110 OPEN 15,8,15, " IOM Open command channel.
120 OPEN 1,8,5, "REL DATA" Open relative file.
210
130 FOR K=l TO NR Loop for all records.
140 H=INT<K/256> Find lo & hi bytes of record number.
150 L=K-256*H
160 PRINT*15,"PnCHR*<5)CHR*<L>CHR*(H)CHR Position to record.
*<1)
170 INPUT#1,A*,B*,C*,D*,E*,F* Use D$ for new index.
180 NI*<K)=D*+RIGHT*<" "+STR*(K>,4> Store image + record #.
190 NEXT K
200 CLOSE 1
• • a •
a • • ■
Sort Nl« using your -favorite sort
a • m m
■ ■ ■ a
300 OPEN 2,8,4, "NEW INDEX, S,W Open new index file.
310 FOR K=l TO NR Loop for all records.
320 R=VAL<RI6HT*<NI*(K),4>> Extract record number.
330 PRINT«2,R Write index number.
340 NEXT K
350 CL0SE2:CL0SE15
3. What happens when your file is empty?
Don't get so involved in planning for your large data base that you forget to think
about what happens when you first run your program. Remember that none of your
files will even exist.
4. Have you covered all the possibilities?
Careful planning and extensive testing is required to make sure you have considered
all possibilities. Often the possibilities of a new record coming at the very beginning
or the very end of a file are overlooked.
5. How will you handle deletions from the file?
The easiest way to handle deletions is simply to use the edit record function to call up
the record to be deleted, edit it to all blanks and delete that record number from the
indices. If you decide to use this technique, you might change the printout routines to
skip over blank records. The problem with this approach is that if you have many
deletions, the file will gradually fill up with blank records. Another way of handling
deletions is to use the edit function to insert the information for a new member into
the record of the member to be deleted. This, however, assumes a relatively knowl-
edgeable user. A more elegant approach would be to automatically blank the record
to be deleted and save the record number in a file of deleted record numbers. When
you want to add new information to the file, the program should first check to see if
there were any deleted records where it could store the new information before it
added a new record to the file. This is a nice, elegant solution but somewhat tricky to
implement.
REMINDER:
You must always specify the byte position in a POSITION command.
Use the fix for the POSITION command (see Section 8.6) whenever your program
calls for writing or reading records in the relative file in a non-sequential manner.
211
9.9 Typical Application
For each of the chapters on file handling (Chapters 6, 7, 8 and 9), I have developed a
mailing list program to illustrate a practical use for the techniques you have learned.
Only those portions of the program that deal with data handling are changed from
one chapter to the next. The portions relating to data entry, editing and printouts
remain the same.
In this chapter, the mailing list program uses an indexed relative file for storing the
data about the club members. The records are stored in the relative file in the order
in which they are entered. The relative file is indexed alphabetically by last name/first
name and by zip codes. To retrieve a record for editing, you must know its record
number. The record numbers are given on the membership list printout.
Background information:
You've just been elected secretary of a local civic organization. As secretary, you are
responsible for making up mailing labels for the monthly newsletter that goes out to
all 500 members. You would like to use your new computer to take some of the
drudgery out of the task.
Analysis of the problem:
You need a simple mailing list program to keep track of the names and addresses of
the 500 members. The program should allow you to add new members, modify
members' names and addresses, and produce an alphabetized list of members or a set
of mailing labels sorted by zip codes.
Planning a record:
The first step in planning the program is to decide what information we need about
each member. You've checked over the club records and made up the following
summary.
Field
Description
Typical Case
Worst Case
1
2
3
4
5
6
Last name
First name
Street address
City
State
Zip code
8 characters
6 characters
18 characters
8 characters
2 characters
6 characters
14 characters
9 characters
23 characters
18 characters
2 characters
6 characters
Total
48 characters
72 characters
Record organization:
Each record in the relative file will consist of the six fields indicated above. We will
use variable length fields. This allows us to use a somewhat shorter record. We
anticipate that our longest record will contain about 60 characters (the average of 48
and 72). As a result, we will use a record size of 66 (60 characters + 5 field delimiters +
1 record delimiter). A typical record will look like this:
212
10
20
— t—
30
40
— I —
50
— H —
60
Jones*TiiJ>»23 First St.*Anytown*AS*123456»////padded with nulls////
* = carriage return character
One limitation of using variable length fields is that we have to be careful not to
overfill the record. The program must always check the total length of the actual
record before we write it to make sure we don't lose the end of the record.
Indices:
Since we need to be able to produce an alphabetized list and a set of labels sorted by
zip codes, we need to have two index files.
Let's go over the main parts of the program to see how it works:
Line Range Description
1000-1100 Initialize variables and ask user to insert disk.
Open command channel and data file.
1120-1200
1220-1280
1300-1460
1500-1610
1650-1860
1880-1910
1950-2390
2410-2650
2690-2960
3000-3370
Position to record #1 and input the number of records in the file. If
there are no records, branch to menu.
Open the alphabetic index file, read the index into an array, and then
close the file. Repeat for the zip code index.
Display main menu, input option chosen and branch to the appropriate
part of the program.
Increment count of existing members, input the data for the new
member, assemble the record and check it for length. Position to
record #1 to update to record count and then store new record in
its place.
Call routines to sort the new record into the two indices.
Input the number of the record to edit, position to the record and read
it Display the existing data, input any changes, rewrite the record,
and if any changes have, been made in the key fields, re-sort the
record into the indices.
Subroutines to delete the edited record's number from the indices
and then re-sort the record into the indices again. This is needed
because editing may have changed the record's position.
Print a list of members on the screen or printer. The list may be
sorted alphabetically or by zip code.
Print mailing labels on the screen or printer. The labels may be
sorted alphabetically or by zip code. If the user presses Q, the printing
213
is suspended after the next label is printed and the user has the option
of aborting the print job or restarting it at any point. This is helpful
for paper jams!
3410-3720 Close the main relative file, rewrite the two index files and end the
program.
3740-3980 Subroutine to find a record's correct position in the alphabetic index
and insert the record number into the index.
4010-4240 Subroutine to find a record's correct position in the zip code index
and insert the record number into the index.
4260-4320 Subroutine to position to record number "k" in the main relative file
and read the name field.
4350-4400 Subroutine to position to record number "k" in the main relative file
and read the zip code field.
*******************
PROGRAM: MAIL INDEX
*******************
0 REM 1541 USER'S GUIDE SECTION 9.9
1 REM COPYRIGHT: G. NEUFELD. 19G4
2 :
3 REM MAIL-LIST PROGRAM USING INDEXED RE
LATIVE FILE
4 :
lOOO REM* MAILING LIST (INDEXED RELATIVE
)
lOlO DIM AX(500),ZX(500>
1020 R*=CHR*(13):REM CARRIAGE RETURN CHA
RACTER
1030 PRINTCHR*(14)CHR*(8>
1040 PRINT" CCLR> CRVS> MAILING LIST
PROGRAM
1050 PRINT" {DOWN} INSERT DATA DISKETTE IN
DISK DRIVE"
1060 PRINT" <DOWN> PRESS {RVS> RETURN*
ROFF> WHEN READY"
1070 :
1080 GET A*: IF A*<>"" GOTO 1080
1090 GET A*: IF A*OCHR*(13) GOTO 1090
llOO PRINT" OJP> ONE MOMENT PLEASE WHILE
I SET UP"
1110 :
1120 REM OPEN COMMAND CHANNEL & INITIAL I
ZE
1130 OPEN 15,8, 15, "10"
1140 INPUT#15,E,E*,T,S
214
1150 IF E>19 THEN PRINT E; E*; T; S: CLOSE 15
:stop
1160 :
1170 REM OPEN RELATIVE FILE
1180 OPEN 1,8,7, "INX MAIL DATA,L, "+CHR*<
66)
1190 INPUT#15,E,E*,T,S
1200 IF E>19 THEN PRINT E; E*; T; S: CL0SE15
ZSTOP
1210 :
1220 REM POSITION TO RECORD #1
1230 PRINT#15, "P"+CHR* (7) +CHR* ( 1 ) +CHR* <0
> +CHR* ( 1 )
1240 INPUT#15,E,E*,T,S
1250 :
1260 REM INPUT NUMBER OF EXISTING RECORD
S
1270 INPUT#1,A*:IF A*="" OR A*=CHR*<255>
THEN A*="0"
1280 NR=VAL(A*):IF NR=0 GOTO 1500
1290 :
1300 REM INPUT ALPHABETIC INDEX
1310 0PEN2,8,6."ALPHA INDEX, S,R"
1320 INPUT#15,E,E*,T,S
1330 IF E>19 THEN PRINTE;E*;T;S:CLOSE2:C
LOSE 1 : CLOSE 1 5 : STOP
1340 FOR K=l TO NR
1350 : INPUT#2, AX(K)
1360 NEXT
1370 CLOSE 2
1380 :
1390 REM INPUT ZIP CODE INDEX
1400 0PEN2,8,6, "ZIP INDEX, S,R"
1410 INPUT#15,E,E*,T,S
1420 IF E>19 THEN PRINTE;E*;T;S:CL0SE2:C
L0SE1 : CLOSE 15: STOP
1430 FOR K=l TO NR
1440 : INPUT#2, ZX(K)
1450 NEXT
1460 CLOSE 2
1470 :
1480 REM* DISPLAY MENU
1490 :
1500 PRINT" {CLRXDOWNJ CRVS> MAILING LIST
MENU "NR" CLEFT* RECORDS CRVS3 "
1510 PRINT" £DOWN>l.
1520 PRINT"<D0WN>2.
ENTER NEW MEMBER"
EDIT EXISTING RECORD
1530 PRINT" {DOWN* 3.
1540 PRINT"{D0WN>4.
PRINT MEMBERS LIST"
PRINT MAILING LABELS
215
1550 PRINT" {DOWN* 5. TERMINATE PROGRAM"
1560 PRINT" {DOWNXRVSJ WHICH OPTION <1-
5) ■;
1570 BETA*: IF A*=""GOTO 1570
1580 A=ASC(A*>-48
1590 IF A<1 OR A>5 GOTO 1570
1600 PRINTA*" "
1610 ON A GOTO 1650,1950,2690,3000,3420
1620 :
1630 REM: ENTER NEW MEMBER
1640 :
1650 NR=NR+l:REM ADD 1 MEMBER
1660 PRINT" <CLR>tRVS> ENTER DATA FOR ME
MBER #"NR" CLEFT} "
1670 NL*="?": INPUT" {DOWNJLAST NAME:";NL*
1680 NF*="?": INPUT" CDOWNJFIRST NAME:";NF
*
1 690 AD*= " ? " : I NPUT " {DOWN* STREET ADDRESS :
" ; AD*
1700 ci*="?": input" <down>city/town: ";ci*
1710 SA*="?" : INPUT" <DOWN>STATE: " ; SA*
1720 ZP*="?": INPUT" (DOWN* ZIP CODE:"jZP*
1 730 F*=NL*+R*+NF*+R*+AD*+R*+C I *+R*+SA*+
R*+ZP*
1740 IF LEN<F*)<60 GOTO 1780
1750 PRINT" {DOWN* CRVS1ERROR: RECORD IS T
OO LONG":X=NR
1760 FOR K=l TO 5000: NEXT: GOTO 2670
1770 :
1780 PRINT" {DOWNXRVS} ONE MOMENT PLEASE
WHILE I UPDATE"
1790 :
1800 REM POSITION TO RECORD # NR+1
1810 NH = INT(<NR+l)/256)
1820 NL = (NR+1) - 256*NH
1 830 PR I NT# 1 5 , " P " +CHR* < 7 ) +CHR* ( NL ) +CHR* (
NH)+CHR*<1)
1840 PRINT#1,F*
1850 INPUT#15,E,E*,T,S
I860 IF E>19 AND E<>50 THEN PRINT E;E*;T
;S:CL0SE15:ST0P
1870 :
1880 VX=NR
1890 GOSUB 3750: REM INSERT INTO ALPHABE
TIC INDEX
1900 GOSUB 4010: REM INSERT INTO ZIP COD
E INDEX
1910 GOTO 1500
1920 :
216
1930 REM* EDIT EXISTING MEMBER
1940 :
1950 PRINT" CCLR> <D0WN> CRVS> EDIT EXISTIN
G MEMBER {DOWN} "
i960 M*=" RECORD NUMBER <l-"+MID* <STR* <NR
),2)+")"
1970 PRINTM*; " O"
1980 PRINT"<UP>"TAB<LEN<M*)>;: INPUT X*:X
=INT(VAL(X*))
1990 IF X<1 OR X>NR GOTO 1500
2000 :
20 lO REM POSITION TO RECORD #X
2020 :
2030 NH = INT<CX+l)/256)
2040 NL = (X+l) - 256*NH
2050 PR I NT# 1 5 , " P " +CHR* < 7 ) +CHR* ( NL ) +CHR* (
NH)+CHR*(1)
2060 I NPUT# 1 , NL* , NF* , AD* .CI*, S A* , ZP*
2070 PRINT" <CLR>{ DOWN J {RVS> EXISTING DAT
A FOR MEMBER #"X" CLEFT* "
2080 PRINT" {DOWN} LAST NAME: ";NL*
2090 PRINT" <DOWN>FIRST NAME: ";NF*
2100 PRINT" €DOWN>STREET ADDRESS: ";AD*
2110 PRINT" CDOWN> CITY/TOWN: ";CI*
2120 PRINT" CDOWNJSTATE: ";SA*
2130 PRINT" O>0WN> ZIP CODE: ";ZP*
2140 PRINT" {HOME} {DOWN 3> "TAB ( lO) : INPUT
TL*
2150 PRINT" <:D0WN>"TAB<11): INPUT TF*
2160 PRINT" {DOWN} "TAB (15>: INPUT AD*
2170 PRINT" {DOWN} "TAB (10): INPUT CI*
2180 PRINT" CDOWN> "TAB <6>: INPUT SA*
2190 PRINT" <DOWN> "TAB (9): INPUT TP*
2200 PRINT" CDOWNXRVS} ONE MOMENT PLEASE
WHILE I UPDATE"
2210 :
2220 REM POSITION TO RECORD #X
2230 NH = INT((X+l)/256)
2240 NL = (X+l) - 256*NH
2250 :
2260 REM FIX FOR POSITION ERROR
2270 :
2280 PR I NT# 1 5 , " P " +CHR* ( 7 ) +CHR* < NL ) +CHR* (
NH)+CHR*<1>
2290 PRINT#15, "P"+CHR* (7) +CHR* <NL) +CHR* (
NH) +CHR* < 1 >
2300 :
2310 T2 = TI + 30
2320 IF TKT2 GOTO 2320
2330 :
217
2340 F*=TL*+R*+TF*+R*+AD*+R*+C I *+R*+SA*+
R*+TP*
2350 IF LEN(F*)>59 GOTO 1750
2360 PRINT#1,F*
2370 IF TF*ONF* OR TL*ONL* THEN NF*=TF
*:NL*=TL*: GOSUB 2420
2380 IF TP*OZP* THEN ZR$=TP*: GOSUB 2550
2390 GOTO 1500
2400 :
2410 REM* SQUEEZE OUT RECORD FROM ALPHA
INDEX
2420 FOR K=l TO NR:REM FIND OLD POSITION
2430 : IF X=AX<K> GOTO 2450
2440 NEXT
2450 IF K>=NR GOTO 2490
2460 FOR L=K TO NR
2470 : AX(L)=AX<L+1)
2480 NEXT
2490 AX(NR)=0
2500 VX=X
2510 GOSUB 3750: REM INSERT INTO ALPHABE
TIC INDEX
2520 RETURN
2530 :
2540 REM* SQUEEZE OUT RECORD FROM ZIP IN
DEX
2550 FOR K=l TO NR:REM FIND OLD POSITION
2560 : IF X=ZX(K> GOTO 2580
2570 NEXT
2580 IF K>=NR GOTO 2620
2590 FOR L=K TO NR
2600 : ZX(L)=ZX(L+1)
2610 NEXT
2620 ZX(NR>=0
2630 VX=X
2640 GOSUB 4010: REM INSERT INTO ZIP COD
E INDEX
2650 RETURN
2660 :
2670 REM* PRINT LIST OF MEMBERS
2680 :
2690 PRINT" tCLRXDOWN>CRVS> PRINT LIST O
F MEMBERS "
2700 INPUT "{DOWN? OUTPUT TO SCREEN OR PR I
NTER (S/P) S«_EFT 3>";Z*
2710 DN=3:IF LEFT*(Z*, 1)="P" THEN DN=4
2720 INPUT" (DOWN* SORTED BY NAME OR ZIP C
ODE (N/Z) N CLEFT 3>";Z*
218
2730 FL=0:IF LEFT* <Z*, 1)="Z" THEN FL=1
2740 PRINT" CCLR>CRVS>REC# LIST OF M
EMBERS {DOWN} "
2750 OPEN 4,DN
2760 FOR K=i TO NR
2770 IF FL=0 THEN X=AX<K>
27SO IF FL=1 THEN X=ZX<K>
2790 REM POSITION TO RECORD #X
2800 NH = INT((X+i>/256>
2810 NL = (X+l) - 256*NH
2820 PRINT#15, "P"-»-CHR* (7) +CHR* <NL) +CHR* (
NH)+CHR*(1)
2830 INPUT#1 , NL*, NF*, AD*, CI*, SA*, ZP*
2840 PRINT#4, RIGHT* <" "+STR*<X> ,3) ; "
"NF*M "NL*", "AD*
2850 PRINT#4, " "CI*", "SA*", "ZP*
2860 IF INT (K/ 10) OK/ 10 GOTO 2900
2870 PRINT" CDOWN} PRESS {RVS} RETURN CROF
F> FOR MORE"
2880 GET A*: IF A*OCHR*(13) GOTO 2880
2890 PRINT" CCLR>€RVS>REC# LIST OF M
EMBERS CDOWN> "
2900 NEXT
2910 CL0SE4
2920 IF DN=4 GOTO 1500
2930 PRINT" {DOWN* PRESS {RVS> RETURN CROF
F> FOR MENU"
2940 GET A*: IF AtO""' GOTO 2920
2950 GET A*: IF A*OCHR*<13) GOTO 2950
2960 GOTO 1500
2970 :
2980 REM* PRINT MAILING LABELS
2990 :
3000 PRINT" {CLR} {DOWN} CRVS> PRINT MAILIN
G LABELS "
3010 INPUT "{DOWN} OUTPUT TO SCREEN OR PR I
NTER (S/P) S CLEFT 3}";Z*
3020 :
3030 DN=3:IF LEFT*(Z*, 1)="P" THEN DN=4
3040 INPUT" {D0WN}S0RTED BY NAME OR ZIP C
ODE (N/Z) NO.EFT 3}"?Z*
3050 FL=0:IF LEFT*(Z*, 1)="Z" THEN FL=1
3060 PRINT" CCLR} CDOWN} CRVS} PRINTING MAI
LING LABELS "
3070 PRINT" {DOWN} PRESS Q TO SUSPEND PRIN
TING {DOWN}"
3080 OPEN 4,DN
3090 FOR K=l TO NR
3100 IF FL=0 THEN X=AX(K)
31 lO IF FL=1 THEN X=ZX<K>
3120 NH = INT((X+l)/256)
219
3130 NL = (X+l> - 256*NH
3140 PRINT#15, "P"+CHR*<7)+CHR*(NL)+CHR*<
NH>+CHR*<1>
3150 INPUT#1,NL*,NF*,AD*,CI*,SA*,ZP*
3160 PRINT#4,NF*H "NL*","
3170 PRINT#4,AD*",M
3180 PRINT#4,CI«", "SA*% "ZP*
3190 PRINT#4:PRINT#4
3200 GET A*: IF A*="Q" GOTO 3290
3210 NEXT
3220 CL0SE4
3230 IF DN=4 60T0 1500
3240 PRINT" {DOWN} PRESS <RVS> RETURN {ROF
FJ FOR MENU"
3250 BET A*: IF A*<>"" GOTO 3250
3260 GET A*: IF A*OCHR*<13> GOTO 3260
3270 GOTO 1500
32SO :
3290 PRINT" {DOWN}OUTPUT ABORTED"
3300 PRINT" {DOWNJ COMMANDS: A = ABORT"
3310 PRINT" R = RESTART"
3320 GET A*: IF A*="A"THEN CL0SE4ZG0T0 15
OO
3330 IF A*<>"R" GOTO 3320
3340 PRINT" (DOWN* LABEL #"K"HAS JUST BEEN
PRINTED"
3350 PRINT" <DOWN> REST ART LIST AT LABEL #
"K+l
3360 PRINT" €UP} "TAB (23): INPUT K:PRINT
3370 GOTO 3100
3380 :
3390 REM* TERMINATE PROGRAM
3400 :
3410 REM POSITION TO RECORD #1 & STORE N
R
3420 PRINT" {DOWN} {RVS} ONE MOMENT PLEASE
WHILE I TIDY UP!"
3430 PR I NT# 1 5 , " P " +CHR* ( 7 ) +CHR* ( 1 ) +CHR* ( O
) +CHR* < 1 )
3440 PRINT#1,NR
3450 :
3460 REM POSITION TO RECORD #20
3470 NL = 20
3480 NH = O
3490 PR I NT# 1 5 , " P " +CHR* ( 7 ) +CHR* ( NL ) +CHR* (
NH>+CHR*(1>
3500 INPUT#15,E,E*,T,S
3510 IF E>19 AND EO50 THEN PRINT E;E*;T
5S:CLOSE15:STOP
3520 CLOSE 1
3530 :
220
3540 REM WRITE ALPHABETIC INDEX
3550 0PEN2,8„6,"@0:ALPHA INDEX. S,W"
3560 INPUT#15,E,E*,T,S
3570 IF EM9 THEN PRINTE;E*;T;S:CL0SE2:P
R I NT# 1 5 , - I 0 " : G0T03550
3580 FOR K=l TO NR
3590 : PRINT# 2, AX <K>
3600 NEXT
36 lO CLOSE 2
3620 :
3630 REM WRITE ZIP CODE INDEX
3640 OPEN2„8,6, "SO: ZIP INDEX, S,W"
3650 INPUT#15,E,E*„T,S
3660 IF E>19 THEN PRINTE;E«; T;S:CL0SE2:P
R I NT# 1 5 , " I O " : B0T03640
3670 FOR K=l TO NR
3680 : PRINT# 2, ZX(K)
3690 NEXT
3700 CLOSE 2
3710 PR I NTCHR* ( 9 ) CHR* < 1 42 ) " <. CLR> BYE "
3720 CL0SE15:END
3730 :
3740 REM* FIND CORRECT POSITION IN ALPHA
BETIC INDEX
3750 mx=nr:mn=i:oa=o:if NR=1 GOTO 3970
3760 SI*=NL*+" "+NF*:REM MAKE SEARCH IMA
GE
3770 :
3780 K=AX < 1 ) : G0SUB4260: REM* LOWER LIMIT
IMAGE
3790 OU*=ZI*
3800 IF SIUKOU* THEN MX=1:G0T0 3930: REM
BEFORE FIRST
3810 :
3820 K=AX(NR-1):G0SUB4260:REM* UPPER LIM
IT IMAGE
3830 UL*=ZI*
3840 IF SI*>UL* THEN MX=NR:GOTO 3970: REM
AFTER LAST
3850 :
3860 NA=INT < <MX+MN) /2)
3870 IF NA=MN GOTO 3930:REM FOUND THE SP
OT
38SO K=AX(NA):G0SUB4260:REM* FIND AVERAG
E IMAGE
3890 IF SI*<ZI*THENUL*=ZI*:MX=NA:GOTO 38
60
3900 IF SI*=ZI* THEN MN=NA:MX=NA+i:G0T03
860
39 lO IF SI*>ZI*THENOU*=ZI*:MN=NA:GOTO 38
60
221
STEP-1
INSERT INTO INDEX
3920 REM* SHIFT UP
3930 FOR K=NR TO MX
3940 AX(K+1)=AX<K)
3950 NEXT
3960 :
3970 AX<MX)=VX:REM*
3980 RETURN
3990 :
4000 REM* FIND CORRECT POSITION IN ZIP C
ODE INDEX
40 lO MX=NR:MN=l:OA=0:IFNR=l GOTO 4230
4020 SI*=ZP*:REM MAKE SEARCH IMAGE
4030 :
4040 K=ZX<l):GOSUB4340:REM* LOWER LIMIT
IMAGE
4050 OL*=ZI*
4060 IF SI*<OL* THEN MX=l:GOTO 4190: REM
BEFORE FIRST
4070 :
4080 K=ZX(NR-1):G0SUB4340:REM* UPPER LIM
IT IMAGE
4090 OU*=ZI*
4100 IF SI*>OU* THEN MX=NR:GOTO 4230: REM
AFTER LAST
4110 :
4120 NA=INT<<MX+MN)/2)
4130 IF NA=MN GOTO 4190:REM FOUND THE SP
OT
4140 K=ZX<NA>:GOSUB4340:REM* FIND AVERAG
E IMAGE
4150 IF si*<zi*thenul*=zi*:mx=na:goto
20
4160
4120
4170
20
4180
4190
4200
4210
4220
4230
4240
4250
4260
4270
4280
4290
41
IF SI*=ZI* THEN MN=NA:MX=NA+l:GOTO
IF SI*>ZI*THENOU*=ZI*:MN=NA:GOTO 41
REM* SHIFT UP
FOR K=NR TO MX
ZX(K+1)=ZX(K)
NEXT
STEP -1
ZX<MX)=VX
RETURN
REM* INSERT INTO INDEX
REM FIND NAME IMAGE FOR RECORD #K
NH = INTUK+D/256)
NL ■ <K+1) - 256*NH
PR I NT# 1 5 , " P " +CHR* < 7 ) +CHR* ( NL ) +CHR* <
NH)+CHR*(1>
4300 INPUT#1 , LZ*, FZ*, AZ*, CZ*, SZ*, ZZ*
4310 ZI*=LZ*+" "+FZ*
222
4320 RETURN
4330 :
4340 REM FIND ZIP CODE FOR RECORD #K
4350 NH = INT((K+l)/256>
4360 NL = (K+l) - 256*NH
4370 PR I NT# 1 5 , " P " +CHR* < 7 ) +CHR* < NL ) +CHR* (
NH)+CHR*<1)
4380 INPUT#1 , LZ*, FZ*„ AZ*, CZ*, SZ*, ZZ*
4390 ZI*=ZZ*
4400 RETURN
Possible modifications and improvements:
1. Modify the editing routine to allow a record to be recalled by the member's name
rather than by record number.
2. Allow the deletion of records and keep a file of deleted records that is scanned
whenever a new member is added so that the new information displaces obsolete
information in the file rather than expands the file.
3. Modify the label printout routine to provide for selective mailing to all individuals
living within a certain geographic area (having a certain range of zip codes).
223
CHAPTER
10
MACHINE LANGUAGE RLE HANDLING
This chapter takes a look at accessing the disk drive from machine language programs.
Each section includes several practical examples to help you understand the proc-
edure. The reader is assumed to be relatively familiar with 6502 machine language
programming and hexadecimal notation.
Only the Commodore KERNAL ROM routines are used for computer/disk drive
communications. This ensures that the routines you develop will not become obsolete
if a minor ROM change is made.
10.1 Introduction to Machine Language File Handling
For most file-handling applications a BASIC program is the quickest and easiest
solution. However, there are some applications such as spelling checkers or programs
that scan large data bases for keywords, where the speed of a machine language
program is essential. This chapter explains how to open a file, read or write to it, and
close it. In addition, it demonstrates how to send disk commands .such as SCRATCH,
VALIDATE and M-R to the 1541 disk drive and monitor the disk status. The routines
in this chapter are given in PAL-64™ assembler format and are designed for use with
the Commodore 64. In most cases, the routines will also work on a VIC-20 because the
routines use the KERNAL routines that are common to both machines.
10.2 Opening a File
There are three KERNAL ROM routines that you must use when opening a file. They
are SETNAM (sets pointers to the file name), SETLFS (sets the logical file), and
OPEN (actually opens the file). Lef s look at how to use them.
SETNAM [SFFBDJ
This is the first routine to use. It sets the pointers to the file name for the OPEN
routine.
Steps in using SETNAM:
1. Load .A with the length of the file name.
2. Load .X with the lo byte of the address of the file name.
3. Load .Y with the hi byte of the address of the file name.
4. JSR to SETNAM routine at $FFBD.
Registers affected: None
225
NOTE:
Even if you are not using a file name (as in OPEN 15,8,15), you still must use this
routine. In this case, load .A with #$00 and JSR to the routine (.X and .Y can have any
values).
EXAMPLE: File name is BUNCHADATA,S,W (len=14) stored at $033C.
LDA #*OE ; LENGTH
LDX #*3C ;lo BYTE POINTER
LDY #*03 JHI BYTE POINTER
JSR *FFBD ;CALL ROUTINE
SERFS (SFFBA)
Use this routine second. It sets the file#, the device# and the channels for use by the
OPEN routine.
Steps in using SETLFS:
1. Load .A with the file# (logical file number).
2. Load .X with the device# (usually eight for disk).
3. Load .Y with the channel# (secondary address).
4. JSR to SETLFS routine at $FFBA.
Registers affected: None
EXAMPLE: We want to use file# 1, device* 8, channels 5.
LDA #«01 ;FILE#
LDX #*OS ; DEVICE*
LDY #*05 ; CHANNEL*
JSR *FFBA ;CALL ROUTINE
OPEN (SFFCO)
This last routine actually opens the file.
Steps in using OPEN:
1. Use SETNAM and SETLFS to set parameters.
2. JSR to OPEN routine at $FFC0.
Registers affected: .A, .X, and .Y
EXAMPLE:
JSR *FFCO 5 OPEN THE FILE
226
ANNOTATED EXAMPLES:
1. OPEN 15,8,15
LDA #*00
JSR *FFBD
LDA #*OF
LDX #*08
LDY #*OF
JSR *FFBA
JSR *FFCO
; LENGTH IS 0
;CALL SETNAM
;FILE* 15
SDEVICE# 8
;CHANNEL# 15
;CALL SETLFS
SCALL OPEN
2. OPEN 15,8, 15, "10"
LDA
#*02 :
(LENGTH
LDX
#< IOMSG i
!L0 BYTE POINTER
LDY
#>IOMSG ;
J HI BYTE POINTER
JSR
*FFBD 1
[CALL SETNAM
LDA
#*0F l
!FILE# 15
LDX
#*08 !
!DEVICE# 8
LDY
#*0F
J CHANNEL* 15
JSR
*FFBA i
iCALL SETLFS
JSR
*FFCO
5 CALL OPEN
IOMSG .ASC "10"
3. OPEN l,8,5,MeO:BUNCHASTUFF,P,W"
LDA
#$12
LDX
#<FILNAM
LDY
#>FILNAM
JSR
*FFBD
LDA
#*01
LDX
#*08
LDY
#*05
JSR
*FFBA
JSR
*FFCO
; length is 18 chr
;lo byte pointer
;hi byte pointer
;call setnam
;file# i
; DEVICE* 8
$ CHANNEL* 5
5CALL SETLFS
5CALL OPEN
FILNAM .ASC "@0:BUNCHASTUFF,P, W"
4. OPEN 1,8,5, "#2"
LDA #$02
LDX #< FILNAM
LDY #>FILNAM
JSR *FFBD
LDA #*01
; length is 2 chr
;l0 byte pointer
;hi byte pointer
;call setnam
$FILE* 1
227
LDX #*08 ; DEVICE* 8
LDY #*05 ; CHANNEL* 5
JSR *FFBA ;CALL SETLFS
JSR *FFCO J CALL OPEN
FILNAM .ASC "*2"
10.3 Reading a File
There are three KERNAL ROM routines which you use to read a file. They are
CHKIN (prepare channel for input), CHRIN (reads one character), and CLRCHN
(clears the I/O channel). Let's look at how to use them:
CHKIN (SFFC6)
This is the first routine you use. It designates the specified file# as the input file.
Steps in using CHKIN:
1. Load .X with the desired file#.
2. JSR to CHKIN routine at $FFC6.
Registers affected: .A and .X
EXAMPLE: Prepare to read from file# 1 (opened previously).
LDX #*01 ;file# 1
JSR *FFC6 jCALL CHKIN
CHRIN (SFFCF)
Use this routine to actually read characters from the input file. If CHKIN has not
been used to specify an input file, data is expected from the keyboard. Note that the
value in .X is destroyed. As a result, you cannot use .X as a counter when reading
data.
Steps in using CHRIN:
1. JSR to CHRIN routine at $FFCF.
2. Byte appears in A.
Registers affected: A and .X
EXAMPLE: Read 256 bytes & store them starting at $2000.
LDY #*00 ;SET POINTER
LOOP JSR *FFCF ;CALL CHRIN
STA *2000,Y ; STORE BYTE
I NY INCREMENT POINTER
BNE LOOP ;READ NEXT
228
CLRCHIM (SFFCCJ
Use this routine once you have finished reading the file. It clears (deselects) the
designated file# and restores I/O to normal (keyboard and screen). NOTE: CLRCHN
does not close any files, it merely deselects the channel temporarily. Do not forget to
use this routine. If you forget, you will have I/O problems.
How to use CLRCHN:
JSR to CLRCHN routine at $FFCC.
Registers affected: .A and .X
EXAMPLE:
JSR *FFCC
;CALL CLRCHN TO RESTORE I/O
ANNOTATED EXAMPLES:
1. Read 256 bytes from file# 3 and store them starting at $4280.
LDX #*03
JSR *FFC6
LDY #*00
LOOP JSR *FFCF
STA $4280, Y
I NY
BNE LOOP
JSR *FFCC
;FILE# 3
;CALL CHKIN
;SET POINTER
J CALL CHRIN
; STORE BYTE
; INCREMENT POINTER
5 READ NEXT
;CALL CLRCHN TO RESTORE I/O
2. Read from file# 5 until carriage return and store them from $033C.
LOOP
DONE
LDX
#*05
;file* 5
JSR
*FFC6
;CALL CHKIN
LDY
#*00
5 SET POINTER
JSR
*FFCF
;CALL CHRIN
CMP
#*OD
; CHECK FOR CARRIAGE RETURN
BEQ
DONE
; BRANCH IF CARRIAGE RETURN
STA
*033C,Y
; STORE BYTE
I NY
INCREMENT POINTER
BNE
LOOP
;read NEXT
JSR
♦FFCC
;CALL CLRCHN TO RESTORE I/O
10.4 Writing a File
There are three KERNAL ROM routines that you use to write to a file. They are
CHKOUT (prepare the channel for output), CHROUT (writes one character), and
CLRCHN (clears the 1/0 channel). Let's look at how to use them:
229
CHKOUT ($FFC9)
This is the first routine you use. It designates the specified file# as the output file.
Note that you must have opened the file previously.
Steps in using CHKOUT:
1. Load .X with the desired file#.
2. JSR to CHKOUT routine at $FFC9.
Registers affected: .A and .X
EXAMPLE: Prepare to write to file# 1 (opened previously).
LDX #*01 ;file# 1
JSR *FFC9 ;CALL CHKOUT
CHROUT(SFFD2)
Use this routine to actually write characters to the output file. If CHKOUT has not
been used to specify an output file, data is sent to the screen. Note that the value in .X
is destroyed. As a result, you cannot use .X as a counter when writing data.
Steps in using CHROUT:
1. Load .A with byte to output.
2. JSR to CHROUT routine at $FFD2.
Registers affected: .A and .X
EXAMPLE: Write out the 256 bytes starting at $2000.
LDY #*00 J SET POINTER
LOOP LDA *2000,Y ;LOAD BYTE
JSR *FFD2 SCALL CHROUT
INY INCREMENT POINTER
BNE LOOP ;REAO NEXT
CLRCHN (SFFCCJ
Use this routine once you have finished reading the file. It clears (deselects) the
designated file# and restores I/O to normal (keyboard and screen). Remember,
CLHCHN does not close any files, it merely deselects them temporarily. Do not forget
to use this routine. If you forget, you will have I/O problems. When you are finished
writing to the file, remember to close it
Steps in using CLRCHN:
1. JSR to CLRCHN routine at $FFCC.
Registers affected: .A and .X
230
EXAMPLE:
JSR *FFCC ;CALL CLRCHN TO RESTORE I/O
ANNOTATED EXAMPLES:
1. Write the 256 bytes starting at $4280 to file# 3.
LDX #*03
JSR *FFC9
LDY #*00
LOOP LDA *4280,Y
JSR *FFD2
I NY
BNE LOOP
JSR *FFCC
FILE# 3
CALL CHKOUT
SET POINTER
LOAD BYTE
CALL CHROUT
INCREMENT POINTER
READ NEXT
CALL CLRCHN TO RESTORE I/O
2. Write bytes starting at $033C until $00 byte to file# 5.
LDX #*05
JSR *FFC9
LDY #*00
LOOP LDA *033C,Y
BEQ DONE
JSR *FFD2
I NY
BNE LOOP
DONE JSR *FFCC
;FILE# 5
;CALL CHKOUT
;SET POINTER
;load BYTE
; BRANCH IF *00 BYTE
;CALL CHROUT
INCREMENT POINTER
5 WRITE NEXT
;CALL CLRCHN TO RESTORE I/O
10.5 Closing a File
Once you have finished reading or writing a file, you must close it. Although there are
two KERNAL ROM routines that appear to be associated with closing a file, CLALL
(close all files) and CLOSE (close one file), you should only use CLOSE.
WARNING:
Do not use CLALL to close disk files. The Commodore 64 Programmer's Reference
Guide seems to imply that this routine can be used to close all open files. However, it
appears that this routine merely works internally. The files are shut down as far as
the Commodore 64 is concerned, but the files are not closed properly on the 1541 disk
drive. As a result, do not use CLALL with disk filed
CLOSE (SFFC3)
This is the routine to use. It closes the file whose file# you specify. Note that you must
have used CLRCHN to clear the file# previously.
Steps in using CLOSE:
1. Load A with the desired file#.
2. JSR to CLOSE routine at $FFC3.
Registers affected: A, .X and .Y
231
EXAMPLE: Prepare to close to file# 1 (cleared previously).
LDX #*OI ;FILE# 1
JSR *FFC3 ;CALL CLOSE
ANNOTATED EXAMPLES:
1. File# 4 was designated output file. Clear and close it.
JSR *FFCC SCALL CLRCHN TO RESTORE I/O
LDA #*04 ;FILE# 4
JSR *FFC3 ;CALL CLOSE
2. Close command channel (f ile# 15).
LDA #*OF ;FILE# IS
JSR *FFC3 ;CALL CLOSE
10.6 Monitoring I/O Status
You may want to monitor the I/O status when opening a file or reading it. Its main
purpose is to detect a device not present or an end-of-f ile condition. In BASIC this is
done by checking the ST variable. In a machine language program you use the
KERNAL ROM routine READST to monitor the I/O status.
READST (SFFE7)
This is the routine to use. It returns the current I/O status in the accumulator. The
meaning of the values returned are summarized below. They are the same as those
for the ST variable in BASIC.
Value Hex Meaning
0
$00
Everything is OK.
1
$01
Time out on write.
2
$02
Time out on read.
64
$40
End of file.
128
$80
Device not present.
Steps in using READST:
1. JSR to READST routine at $FFB7.
2. Interpret code and take action.
Registers affected: .A
232
EXAMPLE: Read current I/O status and branch if EOF.
JSR *FFB7
CMP #*40
BEQ DONE
;CALL READST
; COMPARE TO EOF
5 BRANCH IF EOF
ANNOTATED EXAMPLES:
1. Open command channel. Abort if device not present.
LDA #*00
; LENGTH IS 0
JSR *FFBD
;CALL SETNAM
LDA **OF
;FILE# 15
LDX #*08
j DEVICE* 8
LDY #*OF
; CHANNEL* IS
JSR SFFBA
;CALL SETLFS
JSR *FFCO
;CALL OPEN
JSR *FFB7
jCALL READST
CMP #*80
; COMPARE TO DEVICE NOT PRESENT
BEQ ABORT
; BRANCH IF DEVICE NOT PRESENT
2. Read file to memory starting at $2000. Quit at end of file.
LDA
#*11
LDX
#<FILNAM
LDY
#>FILNAM
JSR
*FFBD
LDA
#*01
LDX
#$08
LDY
**05
JSR
*FFBA
JSR
*FFCO
LDX
#$03
JSR
*FFC6
LDA
#*20
LDX
#*00
STX
*FA
STA
*FB
LDY
#$00
LOOP JSR
*FFCF
STA
(*FA),Y
I NY
BNE
STATUS
INC
*FB
STATUS JSR
*FFB7
CMP
#*40
BNE
LOOP
JSR
*FFCC
LENGTH IS 17 CHR
LO BYTE POINTER
HI BYTE POINTER
CALL SETNAM
FILE* 1
DEVICE* 8
CHANNEL* 5
CALL SETLFS
CALL OPEN
FILE* 3
CALL CHKIN
HI BYTE OF WHERE TO STORE
LO BYTE OF WHERE TO STORE
MAKE O PAGE INDIRECT POINTER
SET INDEX
CALL CHR IN
STORE BYTE USING IND POINTER
INCREMENT INDEX
NOT DONE PAGE
INCREMENT HI BYTE OF POINTER
READ STATUS
IS IT EOF
NO, DO SOME MORE
CALL CLRCHN TO RESTORE I/O
FILNAM .ASC "O: BUNCH ASTUFF,P,R"
233
NOTE:
Another way to monitor the I/O status is to monitor the value stored in ST ($0090) of
the Commodore 64's RAM directly. The FAST COPY program at the end of this
chapter uses this technique.
10.7 Monitoring the Disk Drive Status
You may want to monitor the disk drive's error status when sending direct disk
commands, opening a file, reading from a file or writing to one. It allows you to check
for a wide variety of disk errors. In BASIC this is done by inputting the disk status
from the command channel. The process is similar in a machine language program.
There is no special KERN AL ROM routine to do this job for you.
The steps you follow in reading the disk error status are:
1. OPEN the command channel (if not already open).
2. Use CHKIN to select the command channel for input.
3. Use CHRIN to read the error status and message.
4. Use CLRCHN to deselect the command channel.
CAUTION:
Be sure to input the entire disk status message and not just the error code. You don't
want part of the message left in the drive's error buffer to fool you the next time you
check the disk drive status. You can tell that you are at the end of the message when
you reach a carriage return character.
HINT:
For a quick and dirty check, all you need to do is interpret the first character of the
error code. Since error codes less than 20 can be ignored, just test if the first character
is less than two. Remember that the codes are in ASCII format so that a two is really
50 ($32). If you use this technique, don't forget to dump the rest of the message as
indicated above.
ANNOTATED EXAMPLES:
1. Quick and dirty check of first digit (command channel open).
LDX #*OF ;FILE# 15
JSR *FFC6 ;CALL CHKIN TO SELECT CMD CHNL
JSR *FFCF ;CALL CHRIN TO GET FIRST DIGIT
CMP #*32 ; COMPARE TO *2'
BCS ABORT ; ERROR CODE >= 20 SO ABORT
LOOP JSR *FFCF ;CALL CHRIN TO GET NEXT BYTE
CMP #*OD ; REACHED CARRIAGE RETURN YET
BNE LOOP ; STILL TO COME
JSR *FFCC ;CALL CLRCHN TO RESTORE I/O
234
2. Full interpretation of error code (stored in $FA).
jSR *FFCC ;CALL CLRCHN TO RESTORE I/O
LDX #*OF ;FILE# 15
JSR *FFC6 5 CALL CHKIN TO SELECT CMD CHNL
JSR *FFCF ;CALL CHRIN TO GET 1ST DIGIT
AND #*OF 5 CONVERT FROM ASCII TO HEX
TAY ; TEN'S PLACE
JSR *FFCF ;CALL CHRIN TO GET 2ND DIGIT
AND #*OF ; CONVERT FROM ASCII TO HEX
CPY #*00 J IS TEN'S PLACE ZERO
BEQ DONE ;NO TEN'S PLACE, .A HAS ALL
LOOP1 CLC ; CLEAR CARRY FOR ADD
ADC #*OA JADD IN ONE 10
DEY J DECREMENT TEN'S PLACE
BNE L00P1 ; STILL MORE 10' S TO ADD IN
DONE STA SFA ;SAVE ERROR CODE
L00P2 JSR *FFCF ;CALL CHRIN TO GET NEXT BYTE
CMP #*OD 5 REACHED CARRIAGE RETURN YET
BNE L00P2 ; STILL TO COME
JSR *FFCC JCALL CLRCHN TO RESTORE I/O
NOTE:
See the FAST COPY program at the end of this chapter for a routine that does a
quick and dirty check and, if a serious error has occurred, displays the error code and
message on the screen.
10.8 Sending Disk Commands
As your machine language programs become more complex, you will find it necessary
to send disk commands to the drive, such as NEW, SCRATCH, "M-R", and TO".
This is similar to writing to a file except that you select the command channel f ile# as
the output file#.
The steps you follow in sending disk commands are:
1. OPEN the command channel (if not already open).
2. Use CHKOUT to select the command channel for output
3. Use CHROUT to send the disk command.
4. Use CLRCHN to deselect the command channel.
CAUTION:
You should check the disk error status after sending a disk command to be sure that
the command has been executed properly (see Section 10.7).
ANNOTATED EXAMPLES:
1. Send disk command to NEW a diskette (command channel open). Equivalent to
PRINT#15,"N0:DEVELOPMENT#3,D3".
235
LDX #*0F
JSR *FFC9
LDY #*00
LOOP LDA NEWCMD,Y
JSR *FFD2
I NY
CPY #*14
BNE LOOP
JSR *FFCC
;FILE# 15
;CALL CHKOUT
;SET POINTER
;LOAD BYTE OF COMMAND
SCALL CHROUT
INCREMENT POINTER
SHAVE WE FINISHED 20 BYTES YET
;N0, MORE COMMAND TO COME
;CALL CLRCHN TO RESTORE I/O
NEWCMD .ASC " NO : DEVELOPMENT #3,D3"
2. Send disk command to SCRATCH a file (command channel open). Equivalent to
PRINT#15,"S0:BUNCHADATA".
LDX #*0F
JSR *FFC9
LDY #*00
LOOP LDA SCRCMD,Y
JSR *FFD2
I NY
CPY #*0D
BNE LOOP
JSR *FFCC
;FILE# 15
;call chkout
;set pointer
;load byte of command
scall chrout
s increment pointer
shave we finished 13 bytes yet
sno, more command to come
scall clrchn to restore i/o
SCRCMD .ASC "SOrBUNCHADATA"
3. Send disk command to position to record# 12 in relative file. Equivalent to
PRINT#15,"P";CHR$(4);CHR$(0);CHR$(12);CHR$(1).
LDX #*0F
JSR *FFC9
LDY #*00
LOOP LDA POSCMD,Y
JSR *FFD2
I NY
CPY #*05
BNE LOOP
JSR *FFCC
SFILE* 15
SCALL CHKOUT
5 SET POINTER
5 LOAD BYTE OF COMMAND
SCALL CHROUT
S INCREMENT POINTER
SHAVE WE FINISHED 5 BYTES YET
SNO, MORE COMMAND TO COME
SCALL CLRCHN TO RESTORE I/O
POSCMD .ASC "P"
.BYTE *04,*00,*0C,*01
10.9 Typical Application
One of the most common complaints of 1541 disk drive owners is the length of time it
takes to make a copy of an important disk file. Even some of the commercially
available routines are quite slow. As a result, a file copy program is a prime candidate
for conversion to machine language.
236
The FAST COPY program that serves as the culmination of this chapter illustrates
the use of many of the routines we have developed. It includes routines to open files,
read a file, write a file and check the disk status. The file to be copied is read into the
Commodore 64's memory from $0B00 on. Since the entire RAM area from $0B00 to
$9FFF can be used, files up to about 150 blocks long can be copied. Once the file has
been read in and the copy diskette inserted, the contents of RAM are written into a
new file.
The features of the FAST COPY program include:
1. High speed operation (less than 1.5 sec/block).
2. Copying of PRG or SEQ files.
3. Copying of files up to 150 blocks long.
4. Making multiple copies of a file.
Operating instructions:
1. Load and run to make one or more copies of a file.
2. Run the program again to copy other files.
The program listing is a slight modification of the printer output of the PAL™
assembler (line numbers have been added on comment lines). If you use a different
assembler, you may have to make some minor changes. For example, PAL uses
quotation marks around as ASCII string rather than the apostrophe used by the
MAE™ and Commodore assemblers.
CAUTION:
Only the first three bytes of any ASCII message or hex BYTE string are shown in the
6502 CODE column. Be careful not to miss any characters as you type in the listing.
Since the assembler listing is well documented, no overview is provided. Readers who
would like to make use of the FAST COPY program but do not have access to an
assembler may use the MAKE FAST COPY program in Appendix E. This is a
BASIC program that creates a working copy of the FAST COPY program on a
diskette.
PROGRAM: FAST CPY.P
♦♦**»♦♦*»»******»♦*
0 REM 1541 USER'S BUIDE SECTION 10.9
1 REM COPYRIGHT: G. NEUFELD, 1984
2 :
3 REM MACHINE LANGUAGE FILE COPY FOR THE
COMMODORE 64
4 REM PAL ASSEMBLER FILE
5 :
lOOO REM SAMPLE DISK HANDLING
lOlO REM 1541 FILE 'COPY.P'
237
1020 :
1030 REM OPEN 4,4
1040 OPEN 1,8,4, "@0: COPY. B,P,W"
1050 :
1060 SYS 10*4096
1070 ;
1080 j SAMPLE DISK HANDLING - 1541 RE
AD.P
1090 ;
lioo ;
mo ;
INS THE
1120 ;
RITING
1130 ;
KETTE.
1140 ;
1150 ;
CK
1160 ;
COPIES SEQ OR PRG FILES BY READ
FILE INTO RAM <*0B00 UP) THEN W
OUT A NEW FILE ONTO ANOTHER DIS
FEATURES:
-FAST READ - ABOUT 2 SEC/BLO
-FILES UP TO ISO BLOCKS LONG
1170 ;
1180 ;
-ALLOWS MULTIPLE COPIES
1190
1200
1210
1220
1230
1240
1250
1260
1270
1280
1290
1300
1310
1320
*= $0801
.OPT P,01
■
»
? 10 SYS2061
.BYTE *0B,*08.*0A,*00
.BYTE *9E,*32,*30,*36
.BYTE *31,*00,*00,*00
$
; START OF MACHINE LANGUAGE
■ — ______ _____ _ — _ __
START JSR *FFCC ; CLEAR I/O CHANNELS
1330 JSR *FFE7 ; CLOSE ALL FILES
1340 ;
1350 ; PRINT REQUEST FOR FILE NAME
1360 ;
1370 LDY #*00
1380 ;
1390 LOOP1 LDA MSG,Y
1400 JSR *FFD2 ;CHROUT - OUTPUT A BYTE
1410 INY
1420 CPY #ENDMSG-MSG
238
1430 BNE L00P1
1440 ;
1450 LDY #*00
1460 LOOP2 JSR *FFCF ;CHRIN - INPUT A BY
TE
1470 CMP #*0D ;a CARRIAGE RETURN
1480 BEQ NAME IN
1490 ;
1500 STA *033C,Y ; STORE IN TAPE BUFFER
1510 INY
1520 BNE LOOP2
1530 ;
1540 ; PRINT ERROR MESSAGE
1550 ;
1560 ERROR LDY #*00
1570 LOOP3 LDA ERRMSG,Y
15SO JSR *FFD2 ;CHROUT - OUTPUT A BYTE
1590 INY
1600 CPY #ENDERR-ERRMSG
1610 BNE LOOP3
1620 RTS
1630 ;
1640 NAME IN STY $0334 ;SAVE NAME LENGTH
1650 CPY #*11
1660 BCS ERROR
1670 ;
1680 ; PRINT REQUEST FOR FILE TYPE
1690 ;
1700 AGAIN LDY #*00
1710 5
1720 LOOP4 LDA MSG2,Y
1730 JSR *FFD2 5CHR0UT - OUTPUT A BYTE
1740 INY
1750 CPY #ENDMSG2-MSG2
1760 BNE L00P4
1770 ;
1780 JSR *FFCF ;CHRIN - INPUT A BYTE
1790 STA *0335 ; STORE BELOW BUFFER
1800 CMP #*53 ;IS IT 'S*
1810 BEQ OKTYPE
1820 CMP #*50 ;IS IT 'P'
1830 BNE AGAIN
1840 ;
1850 OKTYPE LDY *0334 ; NAME LENGTH
1860 LDA #*2C ; M,"
1870 STA *033C,Y
1880 INY
1890 LDA *0335 ; FILE TYPE (S/P)
1900 STA *033C,Y
1910 INY
239
1920 LDA #*2C ; ",-
1930 STA *033C,Y
1940 I NY
1950 LDA #*52 ; "RM
I960 STA *033C,Y
1970 I NY
1980 STY *0334
1990 ;
2000 ; OPEN COMMAND CHANNEL
20 10 ;
2020 JSR *FFCC jCLRCHN - CLEAR I/O CHANN
ELS
2030 JSR OPEN 15
2040 ;
2050 ; OPEN FILE TO READ (1,8,5)
2060 ;
2070 JSR OPEN1
2080 ;
2090 JSR STATUS
2100 JSR *FFCC ;CLRCHN - CLEAR CHANNELS
2110 ;
2120 LDA *FB
2130 BEQ CONT
2140 ;
2150 ; ABORT ON BAD DISK STATUS
2160 ;
2170 ABORT LDA #*01 ; CLOSE FILE# 1
2 ISO JSR *FFC3 ; CLOSE - CLOSE A FILE
2190 LDA #*OF ; CLOSE FILE# 15
2200 JSR *FFC3 ; CLOSE - CLOSE A FILE
2210 RTS
2220 ;
2230 ; SET UP INDIRECT POINTER
2240 ;
2250 CONT LDA #*00 ;L0 BYTE START
2260 STA *FC
2270 LDA #*0B ;HI BYTE START
2280 STA *FD
2290 5
2300 ; PRINT LOADING MESSAGE
2310 ;
2320 LDY #*00
2330 LOOP5 LDA RDMSG,Y
2340 JSR *FFD2 ;CHROUT - OUTPUT A BYTE
2350 I NY
2360 CPY #ENDRD-RDMSG
2370 BNE LOOPS
2380 ;
2390 ; READ IN THE FILE
2400 ;
240
2410 LDX #*01 5 FILE*
2420 JSR *FFC6 ;CHKIN - PREPARE FOR INPU
T
2430 »
2440 LDY #*00
2450 LOOP6 JSR *FFCF ;GET BYTE
2460 ST A (*OOFC),Y 5 STORE IN
2470 ;
2480 LDA *90 5KERNAL I/O STATUS FLAG
2490 BNE ENDLOAD
2500 5
2510 INY
2520 BNE LOOP6
2530 INC *FD
2540 BNE LOOP6 ? ALWAYS
2550 ;
2560 ENDLOAD STY *FC ;LAST BYTE POINTER
2570 JSR *FFCC ;CLRCHN - CLEAR CHANNELS
2580 LDA #*01 ; CLOSE FILE# 1
2590 JSR *FFC3 ; CLOSE - CLOSE A FILE
2600 LDA #*OF 5 CLOSE FILE# 15
26 lO JSR *FFC3 ; CLOSE - CLOSE A FILE
2620 i
2630 S PRINT CHANGE DISK MESSAGE
2640 5
2650 CHANGE LDY #*00
2660 LOOP7 LDA CNGMSG,Y
2670 JSR *FFD2 ;CHROUT - OUTPUT A BYTE
26SO INY
2690 CPY #ENDCNG-CNGMSG
2700 BNE L00P7
27 lO ;
2720 ; WAIT FOR CARRIAGE RETURN
2730 ;
2740 WAIT JSR *FFE4; GET IN - GET A CHR
2750 CMP #*0D
2760 BNE WAIT
2770 ;
2780 j OPEN COMMAND CHANNEL
2790 ;
2800 JSR OPEN 15
2810 5
2820 ; OPEN FILE FOR WRITE
2830 ;
2840 LDY *0334 ;NAME LENGTH
2850 LDA #*57 ; "W"
2860 STA *033B,Y
2870 ?
2880 JSR 0PEN1
2890 5
241
2900
2910
2920
2930
2940
2950
2960
2970
29SO
2990
3000
3010
3020
3030
3040
305O
3060
3070
3080
3090
3100
3110
3120
3130
31 40
3150
3160
3170
3180
PUT
3190
3200
3210
3220
3230
3240
3250
3260
3270
3280
3290
3300
3310
3320
3330
3340
3350
3360
LS
3370
JSR STATUS
JSR *FFCC ;CLRCHN
CLEAR CHANNELS
; ABORT ON BAD DISK STATUS
5
LDA *FB
BNE ABORT ; ABORT ON BAD STATUS
5 SET UP INDIRECT POINTER
•
LDA #$00 ;L0 BYTE START
STA *FA
LDA #*OB SHI BYTE START
STA *FB
5
; PRINT WRITING MESSAGE
;
LDY #*00
L00P8 LDA WRTMSG,Y
JSR *FFD2 SCHROUT - OUTPUT A BYTE
I NY
CPY #ENDWRT-WRTMSG
BNE L00P8
»
; WRITE OUT THE FILE
LDX #*01 ;FILE#
JSR *FFC9 5CHK0UT - PREPARE FOR OUT
LDY #*00
LOOPIO LDA <*FA>,Y
JSR *FFD2 SCHROUT - OUTPUT A BYTE
CPY *FC 5 DOES .Y MATCH END COUNT
BNE NOEND
5
LDA *FB
CMP *FD
BEQ ALLDONE
5
NOEND I NY
BNE LOOPIO
s
INC *FB
BNE LOOPIO 5 ALWAYS
ALLDONE JSR *FFCC ; CLEAR I/O CHANNE
LDA #*01 ; CLOSE FILE# 1
242
3380 JSR *FFC3 ; CLOSE - CLOSE A FILE
3390 LDA #*0F ; CLOSE FILE# 15
3400 JSR *FFC3 ; CLOSE - CLOSE A FILE
3410 ;
3420 ; PRINT ANOTHER READ MESSAGE
3430 ;
3440 LDY #*00
3450 LOOP 11 LDA DUNMSG,Y
3460 JSR *FFD2 ;CHROUT - OUTPUT A BYTE
3470 I NY
3480 CPY #ENDDUN-DUNMSG
3490 BNE LOOP 11
3500 »
3510 JSR *FFCC ; CLEAR I/O CHANNELS
3520 ;
3530 ; WAIT FOR Y/N RESPONSE
3540 5
3550 WAIT2 JSR *FFE4; GETIN - GET A CHR
3560 CMP #*59 ; "V"
3570 BEQ CNG
3580 CMP #*4E ; "N"
3590 BNE WAIT2
3600 ;
36 lO JSR *FFCC ; CLEAR I/O CHANNELS
3620 RTS ;BACK TO BASIC
3630 ;
3640 CNG JMP CHANGE
3650
3660
3670
3680
3690
3700
3710
3720 STATUS JSR *FFCC ;CLRCHN - CLEAR CH
ANNELS
3730 ;
3740 LDX #*OF ;FILE#
3750 JSR *FFC6 ;CHKIN - PREPARE FOR INPU
T
3760 ;
3770 JSR *FFCF ; CHR IN - GET BYTE
3780 CMP #*32
3790 BCC DISKOK ;DS<20
3800 ;
3810 STA *FB i TEMP STORAGE
3820 ;
3830 ; PRINT ERROR MESSAGE
3840 ;
3850 LDY #*00
* SUBROUTINES *
SUB TO CHECK DISK STATUS
243
3860 LOOP 12 LDA DSKERR,Y
3870 JSR *FFD2 ;CHROUT - OUTPUT A BYTE
3880 I NY
3890 CPY #ENDDSK-DSKERR
3900 BNE LOOP 12
3910 ;
3920 LDA *FB ; RETRIEVE 1ST DIGIT
3930 JSR *FFD2 ;CHROUT - OUTPUT A BYTE
3940 ;
3950 LOOP13 JSR *FFCF ;CHRIN - GET BYTE
3960 CMP #*OD
3970 BEQ QUIT
3980 JSR *FFD2 ;CHROUT - OUTPUT A BYTE
3990 BNE LOOP 13
4000 BEQ LOOP 13
4010 ;
4020 QUIT LDA #*01 ;SET FLAG
4030 STA *FB
4040 RTS
4050 ;
4060 DISKOK JSR *FFCF ;CHRIN - GET BYTE
4070 CMP #*0D
4080 BEQ QUIT2
4090 BNE DISKOK
4100 BEQ DISKOK
4HO ;
4120 QUIT2 LDA #*00 ; SET FLAG
4130 STA *FB
4140 RTS
4150 ;
4160 ; OPEN COMMAND CHANNEL (OPEN 15,8,1
5, "IO")
4170 ;
4180 OPEN 15 LDA *02 ;NAME LENGTH
4190 LDX #<I0MSG ;LO OF FILE NAME
4200 LDY #>IOMSG ;HI OF FILE NAME
42 IO JSR *FFBD ? SETNAM - SET UP FILE NAM
E
4220 ;
4230 LDA #*0F ;FILE#
4240 LDX #*08 ;DEVICE#
4250 LDY #*0F ; CHANNEL*
4260 JSR *FFBA ;SETLFS - SET UP LOGICAL
FILE
4270 ;
4280 JSR *FFCO ; OPEN COMMAND CHANNEL
4290 ;
4300 RTS
4310 ;
4320 ; OPEN FILE TO READ <1,8,5)
244
4330 ;
4340 0PEN1 LDA *0334 ;NAME LENGTH
4350 LDX #<*033C ;L0 BYTE OF START
4360 LDY #>*033C ;HI BYTE OF START
4370 JSR *FFBD ;SETNAM - SET UP FILE NAM
E
4380 ;
4390 LDA #*01 ;FILE#
4400 LDX #*08 ;DEVICE#
44 lO LDY #*05 ; CHANNEL*
4420 JSR *FFBA ;SETLFS - SET UP LOGICAL
FILE
4430 5
4440 JSR *FFCO ;OPEN FILE
4450 ;
4460 RTS
4470 ;
4480 5 *=======================-= — =
4490 ; * MESSAGES
4500 ; ♦=============—=—==—===———
4510 ;
4520 MS6 .BYTE *8E ; UPPER CASE
4530 .BYTE *93,*11 ; CLEAR SCR, DOWN
4540 .ASC "FILENAME TO COPY? "
4550 ENDMSG =*
4560 ;
4570 ERRMSG .BYTE *0D,*0D,*12
4580 .ASC "FILENAME TOO LONG"
4590 .BYTE *0D
4600 ENDERR =*
4610 ;
4620 MSG2 .BYTE *OD.*OD
4630 .ASC "FILE TYPE <S/P>? "
4640 ENDMSG2 =*
4650 ;
4660 DSKERR .BYTE *0D,*0D,*12
4670 .ASC "FATAL DISK ERROR"
4680 .BYTE *0D
4690 ENDDSK =*
4700 ;
4710 IOMSG .ASC "10"
4720 ;
4730 RDMSG .BYTE *OD,*OD
4740 .ASC "READING DISK FILE"
4750 -BYTE *OD
4760 ENDRD =*
4770 ;
245
4780 CNGMSG .BYTE *OD
4790 -ASC "INSERT OTHER DISKETTE"
4800 .BYTE *OD,*OD
4810 .ASC "PRESS "
4820 .BYTE *12 ;RVS ON
4830 -ASC "RETURN"
4840 .BYTE *92 ;RVS OFF
4850 .ASC " WHEN READY"
4860 ENDCNG =*
4870 ;
4880 WRTMSG .BYTE *0D,*91
4890 -ASC "WRITING DISK FILE
4900 .BYTE SOD
49 lO ENDWRT =*
4920 ;
4930 DUNMSG .BYTE *OD
4940 -ASC "ANOTHER COPY <Y/N>?"
4950 .BYTE *0D
4960 ENDDUN =*
Possible modifications and improvements:
1. To copy very long files (up to 154 blocks) assemble the machine code portion to
begin at $C000 and modify the starting value of the indirect pointers to point to
$0800.
2. To handle exceptionally long files (up to almost 200 blocks) incorporate the
KERNAL routines into the copy program and then flip the BASIC ROM's out to
provide a clear RAM area from the end of the extended routine to $CFFF.
246
CHAPTER
11
GETTING OUT OF TROUBLE
Despite your most careful efforts to follow proper operating procedures when using
your 1541 disk drive, Murphy's Law will eventually catch up with you. This chapter
is designed to help you localize and correct most of the common problems.
The chapter includes listings of several disk utility programs. In addition it makes
extensive use of some of the utility programs listed in Appendix E.
11.1 Spotting and Diagnosing Problems
One of the difficulties of working with floppy diskettes is that errors can develop in
the directory or BAM without your knowledge. If these errors are not found and
corrected, they can precipitate other problems. In accordance with Murphy's Law,
problems on a diskette will not become evident until you desperately need the pro-
grams or data stored on it. This section describes how to use two of the disk utility
programs listed in Appendix E to check for and locate errors hidden on your diskettes.
The program FULL DIRECTORY, listed in Appendix E, provides a rapid, although
incomplete, check for potential problems. The program is very easy to use. Load it,
insert the diskette to be checked and run the program.
Here's an analysis of the 1541TEST/DEMO diskette.
DISK NAME: 1541 TEST/DEMO
DOS TYPE: 65
COSMETIC DISK ID: ZX ASCII = 90
88
BLKS
FILE NAME
TYPE LINK START
13 HOW TO USE PR6 17 O *0401
5 HOW PART TWO PR6 17 3 $0401
4 VIC-20 WEDGE PRG 17 9 *0401
1 C-64 WEDGE PRG 19 0 *0401
4 DOS 5. 1 PRG 19 1 *CCOO
11 COPY/ALL PRG 19 3 *0401
9 PRINTER TEST PRG 19 9 *0401
247
4
DISK ADDR CHANGE
PRG
16
0
*0401
4
DIR
PRG
16
1
*0401
6
VIEW BAM
PRG
16
3
*0401
4
CHECK DISK
PRG
16
7
*0401
14
DISPLAY T&S
PRG
16
15
*0401
9
PERFORMANCE
TEST
PRG
20
2
*0401
5
SEQUENTIAL FILE
PRG
20
7
$0401
13
RANDOM FILE
PRG
15
1
*0401
BLOCKS FREE PER DIRECTORY = 55S
BLOCKS FREE ACCORDING TO BAM= 558
SCRATCHED BLOCKS= O
This is a much more detailed view of a diskette's directory than you get from an
ordinary directory listing. An analysis is provided for all files, even scratched files.
Let's spend a moment going over the table of files and the meaning of the various
entries. The first column of the table, BLKS, is the length of the file, in blocks, as
recorded in the directory. The actual length of the file may be different. The second
column lists the file name as recorded in the directory. The third column is the file
type. Scratched file types have a reverse-field DEL in this column. Unclosed files
have their file types in reverse-field. The next two columns are the track and sector
numbers where the first block of the file is stored. The final column is the load
address of the file expressed in hexadecimal notation. On the 1541TEST/DEMO
diskette the load address for all the programs, except DOS 5.1, is $0401 (1025). This
indicates that these programs were written on a PET. The normal load addresses for
some Commodore computers are:
Commodore 64
$0801
(2049)
VIC-20 (unexpanded)
$1000
(4097)
VIC-20 (with 3K)
$0401
(1025)
VIC-20 (with 8/16K)
$1201
(4608)
4032 & 8032
$0401
(1025)
Although the expanded directory listing is interesting, the summary below the table
is more important for spotting trouble. The first line, BLOCKS FREE PER DIR-
ECTORY, shows 558, which is calculated by subtracting the sum of the first column
of the table from the total number of usable blocks on the diskette (664). In our
example, we would have SUM = 13+5+4+1+4+11+9+4+4+6+4+14+9+5+13 = 106. Thus
our BLOCKS FREE PER DIRECTORY is 664 -106 = 558. (Note that we don't have
to worry about the directory on track 18 in our calculations. The 19 sectors on this
track have already been deducted, 683 -19 = 664.) The second summary line, BLOCKS
FREE ACCORDING TO BAM, is the number that normally appears at the bottom
of a directory listing. The third line, SCRATCHED BLOCKS, is the sum of the
lengths of all the scratched files on the diskette. Since there are no scratched files on
the 1541TEST/DEMO diskette, we have zero scratched blocks.
If the two independent calculations of the number of BLOCKS FREE agree, as
above, no diagnostic message is printed. If the number of blocks free according to the
248
BAM is greater than the number calculated from the directory, something is wrong,
and the diagnostic message, CROSSED FILES LIKELY, is printed. If the number
of blocks free according to the BAM is less than the number calculated from the
directory, the message, NON-FILE BLOCKS ALLOCATED, is printed and the
number of such blocks is indicated. If you get either of these diagnostic messages, you
should localize the problem using the CONFIRM ALL FILES program described
below.
Although the FULL DIRECTORY program will not find all diskette problems, it
provides a quick check of your diskette. It is a good idea to keep a printed copy of the
output of this program for each of your diskettes. It makes recovery from problems
such as a corrupted directory much easier.
A more thorough analysis of a diskette is produced by the utility program CONFIRM
ALL FILES. This program is also easy to use. Simply load CONFIRM ALL FILES,
insert the diskette to be analyzed, and run the program. You may direct the output
either to the screen or a printer. You can also choose whether or not to display the file
chains for the files or the SUPER BAM.
This program traces through all the active files stored on a diskette looking for
errors. It even gives you the option of tracing scratched files. To see how this program
works, let's look at the first part of an analysis of the 1541TEST/DEMO diskette.
CONFIRMING 1541TEST/DEM0 ZX 2A
CHECKING DISKETTE BAM
BAM IS INTERNALLY CONSISTENT
TRACING: HOW TO USE PRG
17.00 17,10 17,20 17,OS 17,18 17,06
17,16 17,04 17,14 17,02 17,12 17,01
17,11
13 BLOCKS FOUND, 13 IN DIRECTORY
TRACING: HOW PART TWO PRG
17,03 17,13 17, OS 17,15 17,07
5 BLOCKS FOUND, 5 IN DIRECTORY
TRACING: VIC-20 WEDGE PRG
17,09 17,19 17,17 16,05
4 BLOCKS FOUND, 4 IN DIRECTORY
TRACING: C-64 WEDGE PRG
19,00
1 BLOCKS FOUND, 1 IN DIRECTORY
TRACING: DOS 5.1 PRG
19.01 19,11 19,02 19,12
4 BLOCKS FOUND, 4 IN DIRECTORY
249
Analysis of remaining programs
ALL FILES HAVE BEEN CHECKED
Let's examine what happens as a diskette is analyzed and what the printout means.
When the analysis begins, the diskette's name and cosmetic diskette ID are printed
out for identification purposes. The message, CHECKING DISKETTE BAM,
indicates that the program is now reading and analyzing the diskette's Block
Availability Map. This takes about 30 seconds. Any discrepancies found in the BAM
(i.e., when the master count of blocks free on a track does not match the bit map) are
printed out. If no errors are found, the message, BAM IS INTERNALLY CONSIS-
TENT, is printed and the file analysis begins.
Let's use the first file on the diskette, HOW TO USE, as an example of how a file is
analyzed.
TRACING: HOW TO USE PRG
17,00 17,10 17,20 17, 08 17,18 17,06
17,16 17,04 17,14 17,02 17,12 17,01
17,11
13 BLOCKS FOUND, 13 IN DIRECTORY
The first line of the display is for identification purposes and reports the name and
type of the file as given in the directory.
TRACING: HOW TO USE PRG
(file name) (file type)
The file is then traced by following the track and sector links. If you have requested a
display of the links, the track and sector numbers of each of the blocks that is used to
store the file will be displayed like this:
17,00 17,10 17,20 17,08 17,18 17,06
17,16 17,04 17,14 17,02 17,12 17,01
17,11
In this case, the first block of the file is stored on track 17, sector 0. The second block is
stored on track 17, sector 10, and so on....
As the file is traced the program checks that each block in the file:
1. Can be read without error.
2. Is properly allocated in the diskette's BAM.
3. Is not being used as part of a previously traced file.
If a block in the file cannot be read without error, the program reports the type and
location of the error and goes on to the next file. If a block of the file is not allocated in
250
the BAM, the track and sector numbers for that block are shown in reverse-field. If a
block is also being used as part of a previously traced file, the name of the other file
and the location of the collision are reported and the program goes on to the next file.
Once all the files have been traced, you have the option of displaying the SUPER
BAM of the diskette. The SUPER BAM of the 1541TEST/DEMO diskette looks
like this:
SUPER
BflM
-hH-
OF
19 h
ISh
17h
16 h
15 h
14 h
-H-
13 i I 111)111
12 ill I 1 I 1 i I
11 I i I I I I I 1 I
10 H I 1 I 1 1 i-+
91 I i 1 i i I 1 i
o
-H-
7 I 1 I I I I I 1 I
E
C
T
0
R
MINIMI
4 111 MINI
3111 II II 1 I
21 1 1 1 I 1 I 1 1
1 H-H-B+-H-
@ H I I I I I I I
-HLHfl
-HOKC
HHi.LR4GG-fH-+
-H0KC-FN-H4
-HLJfl-+fH
-KiQLB-FM
-HLJfl-FM
-H00JB-FM4
-H++LIFHEIH
•H-HOIfl-ffiN-t
-H-HLHfl-H3C
++++JKC-H3G-1-
-H-HLHR-FG
•H-HLKB-FN
-H-H.Jn-FMH
-HOOCB^M
+HLJfl@Fn
-HOOJB-Fn
++4LIR-FM-H
HDIRSE
-H-HLHR@BG
l...l..,l-l,.J..,l,.J, .I, I I l,,l„,„l I ..L, .1,1 I.I |„l, l,l.,l„.l l,l„ I ,1.
1 234567890 1 234567890 1 234567890 1 2345
KEV TO SUPER BflM
i3 = IHRECTORV (TRACK 13 ONLV)
fl = HOW TO USE
B - HOW PRRT TWO
C = VIC-20 WEDGE
D = C-64 WEDGE
E = DOS 5. 1
F = COPY/ALL
G = PRINTER TEST
H = DISK RDDR CHANGE
I = DIR
J = VIEW BflM
K = CHECK DISK
L = DISPLAY T&S
M = PERFORMRNCE TEST
251
H = SEQUENTIAL FILE
0 = RANDOM FIAL
The SUPER BAM is a map of the diskette. It shows in a compact form which blocks
are being used to store the various files. For example, if we look at the vertical column
that corresponds to track 17, we see several A's. Checking the key at the bottom of the
SUPER BAM we find that an A represents a block being used to store part of the file
HOW TO USE. If you compare the SUPER BAM with the printout of the file chain
of HOW TO USE, you will see that they agree. Both indicate that the file is stored in
sectors 0, 1, 2, 4, 6, 8, 10, 11, 12, 14, 16, 18, and 20 of track 17. However, the file chain
also gives the sequence in which the sectors are used.
NOTE:
Unallocated blocks which are part of a file and allocated blocks which are not part of
a file, are signs of trouble on the diskette. They are highlighted in reverse-field in the
SUPER BAM printout.
Solving the Problems You Have Found
If any of the blocks are highlighted on the SUPER BAM of a normal diskette, you
should VALIDATE the diskette immediately (see Section 4.8). This should solve the
problem. If you can't VALIDATE the diskette because of one or more read errors,
you will have to copy the files onto another diskette. You can use the FAST COPY
program at the end of Chapter 10 to do this.
Other problems such as crossed files, files containing read errors and files with
incorrect lengths in the directory are symptoms of serious problems on the diskette.
Copy all the files onto another diskette and check the problem files individually.
Hopefully, most of the programs and data will be usable. Do not reuse the problem
diskette until you have completely reformatted it (a full NEW).
WARNING:
Do not remove the write protect tab from a commercially recorded diskette and
validate it! A strange allocation of space on the diskette may be part of a disk protection
scheme. If it is, the program may never run again!
11.2 Unscratching a File
At some point when you are doing diskette housekeeping you will get overzealous and
scratch an important file by mistake. Don't panic. All is not lost. When you scratch a
file, the file is not erased. All that happens is the file type byte in the directory entry
for that file is changed to 00 and the blocks that were used to store the file are marked
as free (available for use) in the diskette's BAM. As long as you have not stored any
new files on that diskette, the original information is still intact.
As soon as you realize that you have scratched a file by mistake, label the diskette so
that you won't use it inadvertently. Now, get out your copy of THE UNSCRATCHER
disk utility program (listed in Appendix E). Load and run THE UNSCRATCHER
program. When you do, you will see a display like this:
252
1541 FILE UNSCRATCHER
INSERT DISKETTE WITH SCRATCHED FILE
PRESS C TO CONTINUE
Q TO QUIT
Insert the diskette with the scratched file and press C. The computer will respond
with:
OK. READING DIRECTORY...
Once the diskette's directory has been read, the name of any scratched files will be
displayed. If your diskette contained three scratched files named OLD STUFF, JUNK
and BUNCHADATA, the display would look like this:
1541 FILE UNSCRATCHER
3 SCRATCHED FILES FOUND
1. OLD STUFF
2. JUNK
3. BUNCHADATA
WHICH FILE TO UNSCRATCH (1-3)?
Suppose you wanted to recover BUNCHADATA. You would type the number three
and press the RETURN key. Once you have selected the file to be unscratched, the
BAM will be read and the file traced to be sure that all the blocks are still free. If
everything is OK, you will be asked if the file is to be unscratched.
1541 FILE UNSCRATCHER
3 SCRATCHED FILES FOUND
1. OLD STUFF
2. JUNK
3. BUNCHADATA
WHICH FILE TO UNSCRATCH <l-3>? 3
READING THE BAM PRIOR TO CHECK
CHECKING IF ALL BLOCKS ARE FREE...
FILE IS 24 BLOCKS LONG AND
CAN BE UNSCRATCHED
SHALL I UNSCRATCH IT <Y/N>?
253
If you respond by typing Y and pressing the RETURN key, you will be asked for the
file type to use.
SHALL I UNSCRATCH IT <Y/N>? Y
FILE TYPE (P/S/U)? _
Enter P for a program file, S for a sequential file or U for a user file. The
UNSCRATCHER program also works with relative (REL) files. If the file you have
selected for unscratching is a relative file, you will not be asked for the file type. It
will be displayed automatically like this:
SHALL I UNSCRATCH IT <Y/N>? Y
THIS IS A RELATIVE FILE
RECORD SIZE - 40
Once the file type has been determined, the file will be unscratched by changing the
file type byte in the directory from zero to the appropriate value (SEQ=129=$81,
PRG=130=$82, USR=131=$83, RELf132=$84) and the diskette validated to allocate
the blocks that make up the file in the BAM.
NOW VALIDATING DISKETTE
ALL DONE!
PRESS RETURN TO CONTINUE
If any disk errors are encountered, the DOS error message will be displayed and the
unscratch job aborted. You may or may not be able to see the file in a directory
listing. However, the blocks of the unscratched file will not be allocated in the BAM
and there is still a danger of overwriting the file if new information is stored on the
diskette. If you can see the entry in a directory listing, all is not lost. Unscratch any
other files you want to recover from the problem diskette now. Once you can see all
the files you want to keep in a directory listing, copy them one at a time from the
problem diskette onto another diskette. Use a copy program such as FAST COPY at
the end of Chapter 10. Do not use a backup program. Do not reuse the problem
diskette until you have completely reformatted it with a full NEW.
11.3 Recovering Data from an Unclosed File
An unclosed file is one that shows up in a directory listing as being zero blocks long
and has an asterisk in front of its file type (*SEQ, *PRG or *USR). It is called an
unclosed file because the file was never closed properly. An unclosed file may be the
result of a power failure, a program that aborted before it finished writing a file, a
full diskette, a flawed diskette, removing the diskette from the 1541 while the file was
still open, or simply forgetting to close the file.
254
Some data will always be lost from an unclosed file because at least the last block is
missing. It was never recorded on the diskette. To help you get a feel for what an
unclosed file is like, let's try some experiments. Get out a test diskette and look at its
directory. Be sure to write down the number of blocks free. Now type in this program
and run it.
10 OPEN 15,8, 15, "I "
20 OPEN 1,8,5, "UNCLOSED, S,W"
30 FOR K=10 TO 59
40 PRINT#1, "THIS IS RECORD #";K
50 NEXT
60 STOP
When the program stops, the red drive-active LED will still be on. Open the door of
the drive and remove the test diskette. The LED will go out.
Now load the test diskette's directory again and list it. There will be a new directory
entry that looks like this:
O "UNCLOSED" *SEQ
Notice that the number of blocks free on the diskette has decreased. Let's calculate
how much space our file should take. Each of our records should occupy a total of 21
bytes of storage:
THIS IS RECORD # 10 *
+ + + + +_
1 5 10 15 20
* —
= carriage return character
Since we have a total of 50 records, we should have used up 50 x 21 = 1050 bytes of
disk storage space. Each block (sector) on the diskette can hold 254 bytes (two bytes
are used for the forward pointer). Therefore our file should occupy a total of 1050/254
= 4.13385827 blocks.
Since the last block was never written to diskette, our file should have used up four
full blocks on the diskette. Even if the next block has already been allocated in the
BAM, we should only use five blocks. Did you find a difference of four or five in the
number of blocks free? Ill bet you didn't! Surprisingly, the actual decrease in the
number of blocks free depends on where the file is stored on the diskette. If you are
using a blank diskette, the number of blocks free will decrease from 664 to 643 (21
blocks). The reason for the discrepancy is the strange way in which the 1541 allocates
blocks. When a file is opened, all the blocks remaining on a track, or if a new track is
needed, the blocks of an entire track, are allocated in the BAM and recorded on track
18, sector 0 of the diskette. The DOS maintains a separate BAM in the drive's RAM
memory. Until the file is properly closed these two versions of the BAM will not be
255
the same. The BAM in RAM will show the correct allocation of blocks while the BAM
on diskette will show an over allocation of blocks. When the file is closed, the correct
BAM is recorded on diskette.
This means that although the BAM of our diskette is incorrect, the blocks that were
used to store the unclosed file are all allocated. As a result, storing new information
on this diskette will not overwrite the unclosed file.
WARNING:
An unclosed program file that was produced by an aborted SAVE is not protected! In
this case, none of the blocks in the file will have been allocated in the diskette's version
of the BAM. Be sure to use the ALLOCATE program to allocate the blocks in such a
file before you use the RECOVER UNCLOSED program.
Recovering the data stored in an unclosed file is a bit tricky. You cannot OPEN an
unclosed file to read it using normal READ (R) mode. You have to use a special
MODIFY (M) mode. Let's use a little test program to see how this works. Insert the
test diskette in your drive, type in this short program and run it.
10 OPEN 15,8, 15, "I"
20 OPEN 1,8,5, "UNCLOSED, S,M"
30 GET#1, A*: PRINT A*;
40 IF STOO THEN PRINT "ST="ST
50 GET B*:IF B*=,,H GOTO 30
60 CLOSE l: CLOSE 15
You should see something like this:
THIS IS RECORD
#10
THIS IS RECORD
#11
THIS IS RECORD
#12
THIS IS RECORD
#56
THIS IS RECORD
#57
THIS IS
(garbage follows)
Press
Press any key to stop!
The first part of the file can be read satisfactorily but the last part is lost. Note that
the status variable, ST, is always zero. No end-of-file marker is present.
Since an unclosed file has no end-of-file marker and reading past the end of the data-
produces garbage, you must monitor what is being read and abort the reading process
once the end of the file is reached. Here are two short programs to assist you in
recovering the data from an unclosed file. The first program, RECOVER
UNCLOSED, is used to recover the data from any type of unclosed file (note that it is
impossible to have an unclosed REL file). Before you use this program to recover a
program file, use the second program, ALLOCATE, to ensure that all the blocks in
the file have been properly allocated.
256
*******************
program: RECOVER UN
*******************
NOTE:
This program is used to recover the data from any type of unclosed file. The data is
read from the unclosed file, displayed on the screen line by line and then stored in a
new file. After each line of data you will be asked if you wish to abort the process. To
bypass the question simply hold down the SHIFT key. To regain control at any point
press the CTRL key.
CAUTION:
Be sure all blocks in the unclosed file are allocated in the BAM before using this
program.
0 REM 1541 USER'S GUIDE SECTION 11.3
1 REM COPYRIGHT: G. NEUFELD, 1984
2 :
3 REM RECOVER DATA FROM AN UNCLOSED FIL
E
4 :
100 REM READ UNCLOSED
110 B*= " " : Z*=CHR* < O > : QT*=CHR* < 34 ) +CHR* ( 2
O)
120 PRINT" CCLR* RECOVER UNCLOSED FILE-
ISO INPUT" {DOWN* NAME OF UNCLOSED FILE";U
N*
140 INPUT "FILE TYPE OF UNCLOSED FILE <P/
S/U)"5UT*
150 IF UT*="P" OR UT*="S" OR UT*="U" GOT
O 170
160 PRINT" CDOWN* ENTER ONLY 'P% "S% OR
'U'":GOTO 140
170 OPEN 15,8, 15, "10"
180 OPEN 1,8,5, "0:"+UN*+","+UT*+",M"
190 GOSUB 490
200 IF FL=1 THEN CLOSE 1 : CLOSE 15: END
210 INPUT" <DOWN> NAME FOR RECOVERED FILE"
5NF*
220 PR I NT "FILE TYPE OF RECOVERED FILE (P
/S/U) "?UT*
230 PRINT "CUP*" TAB (35) 5: INPUT NT*
240 IF NT*="P" OR NT*="S" OR NT*="U" GOT
O 260
250 PRINT" CDOWNJ ENTER ONLY 'P", 'S', OR
*U'":GOTO 220
260 OPEN 2 , 8 , 6 , " O : " +NF*+ " , " +NT*+ " , W "
257
270 GOSUB 490
280 IF FL=1 THEN CLOSE 1: CLOSE 2: CLOSE 1
5: END
290 :
300 PRINT" {DOWN*PRESS CCTRL3 TO ABORT"
310 PRINT" CSHIFT3 FOR CONTINUOUS R
EADCDOWNV
320 PRINT QT*;
330 ZA*=""
340 GET#1,A*:IF A*=B* THEN A*=Z*
350 ZA*=ZA*+A*
360 A=ASC(A*>
370 IF A<> 141 THEN PRINT A*;: IF A=34 TH
EN PRINT QT*;
380 P=PEEK<653>
390 IF P AND 4 GOTO 430
400 IF LEN(ZA*)=255 GOTO 430
410 IF AOO AND A<>13 GOTO 340
420 IF P AND 1 THEN PRINT: GOTO 460
430 PRINT: INPUT "ABORT NOW (Y/N) NtLEFT
3> " ; X*
440 IF X*="Y" THEN CLOSE l:CLOSE2: CLOSE 1
5: END
450 PRINT" CUP* CUP>"
460 PRINT#2, ZA*; : ZZ=FRE <0)
470 PRINT QT*;
480 GOTO 330
490 INPUT#15,EN,EM*„ET,ES
500 IF EN<20 THEN FL=0: RETURN
510 print" cdown} disk error"
520 printen;em*;et;es
530 FL=1
540 RETURN
PROGRAM: ALLOCATE
*****************
NOTE:
This program is used to allocate all the blocks in an unclosed file before using the
RECOVER UNCLOSED program to recover the data in it. To use this program,
simply load and run it. You will need the name of the unclosed file (wild cards are
OK) and its file type.
0 REM 1541 USER'S GUIDE SECTION 11.3
1 REM COPYRIGHT: G. NEUFELD, 1984
258
2 :
3 REM ALLOCATE ALL SECTORS IN A FILE
4 :
lOO REM ALLOCATE UNCLOSED
1 lO B*=" " : Z*=CHR* <0> : QT*=CHR* (34) +CHR* (2
O)
120 PRINT" CCLRJ ALLOCATE UNCLOSED FILE"
130 INPUT" CDOWNJNAME OF UNCLOSED FILE";U
N*
140 INPUT "FILE TYPE OF UNCLOSED FILE <P/
S/U)"?UT*
150 IF UT*="P" OR UT*="S" OR UT*="U" GOT
O 170
160 PRINT" {DOWN* ENTER ONLY *P* , *S', OR
*U*":GOTO 140
170 OPEN 15,8, 15, "IO"
ISO GOSUB 410
190 OPEN 1,8,5, "0:"+UN*+","+UT*+",M"
200 GOSUB 4lb
210 OPEN 2,8,6, "#"
220 GOSUB 410
230 :
240 PRINT#15, "M-R,,CHR*(144)CHR*(2)CHR*<1
)
250 GET#15,A*:SE=ASC<A*+Z*>
260 PRINT#15, "M-R"CHR*(148)CHR*<2)CHR*<1
)
270 GET#15,A*:PT=ASC(A*+Z*>
280 PRINT#15, "Ul:6 O 18"SE
290 GOSUB 410
300 PRINT#15, "B-P:6"PT+1
310 GOSUB 410
320 6ET#2,A*:T=ASC<A*+Z*>
330 GET#2,A*:S=ASC<A*+Z*>
340 NS=20+2*<T>17)+(T>24)+<T>30)
350 IF T<iOR T>35 OR S>NS THEN PR I NT "DON
E":GOTO 450
360 PRINT#15, "B-A:0"T;S
370 INPUT#15,EN,EM*,ET,ES
380 IF EN>19 THEN PR I NT "ALREADY ALLOCATE
D":GOTO 450
390 PRINT#15, "Ul:6 0"T;S
400 GOTO 320
410 input#15,en,em*,et,es
420 if en<20 then return
430 print" tdown> disk error"
440 printen;em*;et;es
450 CLOSE l: CLOSE 2: CLOSE 15
460 END
259
11.4 Recovering from a Short NEW
When you want to get rid of all the files on a diskette so that you can use it for storing
new files, you can use a short NEW disk command. A short NEW command looks
like this:
syntax:
NO: diskette name
EXAMPLES:
OPEN 15,8, 15, "NO:LOTTA DATA"
PR I NT# 1 5 , " N : NE W GAMES "
>N:WORDPROCESSING 6
Note that no diskette ID is specified in the command. If a diskette ID is included in
the command, the disk drive performs a full NEW and rewrites the entire diskette.
It is possible to recover from a short NEW because most of the information on the
diskette is not erased. Only the diskette's BAM and the first directory block (track 18,
sectors 0 and 1) are altered. The program FIX DIRECTORY listed in Appendix E
will automatically recover all the files on your diskette.
To use the FIX DIRECTORY program to recover from a short NEW simply load
and run the program. When you are asked which sector to fix, enter a 1 to signify that
track 18, sector 1 is to be recovered. After this, everything is automatic. The file type
for each file will be assigned as PRG or SEQ on the basis of the load address at the
start of the file. The program also checks for possible REL files. If a file may be part
of a REL file, you will be asked if this is possible. If you respond with a Y to this
question, the program will attempt to reconstruct the relative file.
Once the program has run to completion, all the files on the diskette will have been
recovered. The first eight entries in the directory will be named FILE # 1, FILE # 2,
and so on, and will have to be renamed. However, this should be easy now that you can
load or read the files.
NOTE:
In a few cases, such as the PRG files produced by WordPro™, PaperClip™ and Paint
Magic™, you may have to use the EDIT T&S program (Appendix E) or the MOD
ENTRY program (Section 13.8) to edit the file type byte to restore the correct
file type.
11.5 Repairing a Damaged Directory
One of the most frustrating problems that you can encounter is a diskette with a
damaged directory. You know that the files are there, but you can't get at them.
Despite what you may have read, it is possible to recover a damaged directory. Here's
what to do.
The first thing to do is eliminate your drive as a possible source of the error. If the
problem occurs on just one diskette, your drive is probably OK. If you are having
260
problems with most of your diskettes, your drive is probably out of alignment. Either
have it aligned professionally or, if you are handy, read Chapter 12 and do it yourself.
Once you are sure the problem is caused by the diskette and not the disk drive, you
must determine which directory sector is causing the problem. You can do this by
attempting to load the directory and then reading the disk drive's error status, or you
can use this short program:
PROGRAM: FIND ERRORS
********************
0 REM 1541 USER'S GUIDE SECTION 11.5
1 REM COPYRIGHT: G. NEUFELD, 1984
2 :
3 REM LOCATE READ ERRORS ON THE D I RECTO
RY TRACK
4 :
100 REM FIND ERRORS
110 PRINT" CCLR>FIND ERRORS IN THE DIRECT
ORY"
120 PRINT" {DOWN J INSERT DISKETTE"
130 PRINT" {DOWN} PRESS {RVSJRETURNtROFF*
TO CONTINUE"
140 GET A*: IF A*OCHR*(13) GOTO 140
150 PRINT" CCLR} ERROR ANALYSIS OF TRACK 1
8 {DOWN}"
160 OPEN 15,8, 15, "10"
170 OPEN 1,8,5, "tt"
180 DIM SE<20)
190 SE<l)=0:KT=l:NE=0
200 FOR K=2 TO 19
210 SE(K)=KT
220 KT=KT+3:IF KTM8 THEN KT=KT-17
230 NEXT
240 FOR K=l TO 19
250 PRINT#15, "Ul:5 O 18"SE(K)
260 INPUT#15,EN,EM*,ET.ES*
270 IF EN<20 GOTO 300
280 NE=NE+1
290 PR I NT "SECTOR #"SE(K) "ERROR #"EN" - "
EM*
300 NEXT
310 CLOSE l: CLOSE 15
320 IF NE=0 THEN PRINT" {DOWN} NO ERRORS F
OUND ON TRACK IS"
330 IF NEOO THEN PRINT" {DOWN}" NE" ERRORS
ON TRACK 18"
261
Once you know which sector is causing problems, write down its sector number and
the error code number. Now that you have this information, proceed to the appropriate
section below.
Problems on Sector 0
If track 18, sector 0 has any read errors, you will be unable to initialize the diskette or
access any files on it. This is the first sector in the directory file and contains the
diskette's BAM, disk name and cosmetic ID.
If there is a 22, 23, or 24, READ ERROR on this sector, you can rewrite the sector
using the program REDO 18/0 listed below. You can try using this program to
recover from the less common 20, 21, or 27, READ ERROR as well, but you are
unlikely to succeed. These errors are the result of problems in the header block. If you
can't eliminate the problem with REDO 18/0, you will have to use the program
BACKUP from Appendix E to make a duplicate of the problem diskette. Unlike
most backup programs, this program will not abort when it encounters a read error.
Once the backup is complete, validate the duplicate diskette and all your files should
be intact.
******************
PROGRAM: REDO 18/0
******************
0 REM 1541 USER'S GUIDE SECTION 11.5
1 REM COPYRIGHT: G. NEUFELD, 1984
2 :
3 REM RECOVER FROM ERRORS ON TRACK 18, S
ECTOR O
4 :
100 REM REDO 18, 0
110 FOR K=1T016:B*=B*+CHR*<160):NEXT
120 PRINT" {CLR} REDO TRACK 18, SECTOR O"
130 PRINT" {DOWN} INSERT DISKETTE INTO DR
IVE"
140 PRINT" PRESS {RVS} RETURN {ROFF } TO CO
NTINUE"
ISO GET R*:IF R*OCHR*(13) GOTO 150
l&O OPEN 15,8,15
170 OPEN 1,8,5, "#"
180 PRINT#15, "Ul:5 O 18 O"
190 INPUT#15,EN,EM*,ET,ES
200 PRINT" CCLRJREDO TRACK 18, SECTOR O"
210 PRINT" <DOWN> DISK STATUS WHEN READIN
G 18/0 IS: "
220 PRINT" {DOWN} "EN;EM*;ET;ES
230 CLOSE 1
240 INPUT#15,EN,EM*,ET,ES
250 INPUT" {DOWN} CONTINUE (Y/N) N{LEFT
3} " 5 R*
260 IF R*<>"Y" THEN CLOSE 15: END
270 PRINT" {CLR}REDO TRACK IS, SECTOR O"
262
280 JOB=176:GOSUB 470
290 PRINT" CDOWN> EMBEDDED DISK ID IS: "ID
*
300 PRINT" {DOWN* DISK NAME DESIREDtDOWN*
CCOM-U 16}"
310 PRINT" CUP 2>"TAB(18);: INPUT DN*
320 DN*=LEFT*CDN*+B*,16>
330 PRINT" {DOWN 2> ONE MOMENT PLEASE
II
340 PRINT#15, "M-WCHR* (O) CHR* <3) CHR* (4) C
HR* < 18) CHR* <4> CHR* <65> CHR* <0>
350 FOR K=4 TO 143:PRINT#15, "M-WCHR*(K>
CHR* (3) CHR* ( 1 ) CHR* (O) : NEXT
360 PRINT#15, "M-W"CHR*(144)CHR*(3)CHR*(2
O) DN*; CHR* < 160) CHR* ( 160) ID*
370 PRINT#15, "M-WCHR* (164) CHR* <3> CHR* (3
)CHR*(32)"2A"
380 FOR K=167 TO 170:PRINT#15. "M-WCHR*<
K) CHR* (3) CHR* < 1 ) CHR* < 160) : NEXT
390 FOR K=171 TO 179: PRINT#15, "M-WCHR* <
K) CHR* <3> CHR* ( 1 ) CHR* (O) : NEXT
400 PRINT#15, "M-W"CHR*(180)CHR*<3)CHR*(i
2) "BLOCKS FREE. "
410 FOR K=192 TO 206:PRINT#15, "M-WCHR* (
K) CHR* (3) CHR* ( 1 ) CHR* <32) : NEXT
420 FOR K=207 TO 255: PR I NT# 15, "M-WCHR* (
K) CHR* <3) CHR* ( 1 ) CHR* (O) : NEXT
430 JOB=144:GOSUB 470
440 PRINT" {DOWN} VALIDATING DISKETTE TO
CORRECT BAM"
450 PRINT#15, "V": CLOSE 15:END
460 :
470 PRINT#15, "M-WCHR* (6) CHR* (O) CHR* (2) C
HR*(i8>CHR*(b>
480 PRINT#15, "M-WCHR* (O) CHR* <0) CHR* (DC
HR*<JOB)
490 PRINT#15, "M-R"CHR*(0)CHR*(0)
500 GET# I 5 , A* : A=ASC ( A*+CHR* ( O ) )
510 IF A>127 GOTO 490
520 IF A=l GOTO 570
530 TY=TY+l:IF TY<5 GOTO 470
540 IF J0B=176 THEN PRINT" CDOWN> CAN'T FI
ND TRACK 18":CL0SE15:END
550 PRINT" WILL NOT WRITE CLEANL Y .": CLOSE
15: END
560 PR I NT "SECTOR ZERO. ERROR #"A+18:CL0S
E15:END
570 PRINT#15, "M-R"CHR*<18)CHR*(0)CHR*(2)
580 GET#15,A*:ID*=A*
263
590 GET# 1 5 . A* : I D*= I D*+A*
600 RETURN
Problems on Sector 1
If track 18, sector 1 cannot be read without error, you will be unable to access any
files on the diskette. This is the second block of the directory file and contains the
directory entries for the first eight files on the diskette.
The FIX DIRECTORY program from Appendix E can recover the files if there is a
22, 23, or 24, READ ERROR on this sector. You can try using this same program to
try to recover from the less common 20, 21, or 27, READ ERROR as well. You won't
cause any harm, but you are unlikely to succeed in eliminating the error. These errors
are the result of problems in the header block.
To use the FIX DIRECTORY program to recover from a read error on sector 1,
simply load and run the program. When you are asked which sector to fix, enter a
one. The program will then attempt to read the sector. If a read error is encountered,
you will be informed of the type of error and presented with three choices:
1. Null out the sector.
2. Bridge over the damaged sector.
3. Terminate the program.
Always choose the null out the sector option at least once. If the sector contains a 22,
23, or 24, READ ERROR, this should repair the damaged sector. If the sector contains
a 20, 21, or 27, READ ERROR, this attempt at repair may not work. If it doesn't, you
must use the program BACKUP from Appendix E to make a duplicate of the problem
diskette (you cannot bridge over errors in sector 1). Once the backup is complete
(about 25 minutes), use the FIX DIRECTORY program with the backup diskette as
indicated above.
Once sector 1 can be read without error, it is checked to make sure it is an empty
sector. If any characters, other than nulls, are found, you will be asked if it is OK to
continue. This is done so you have a chance to abort things if you gave the wrong
sector number.
After this, everything is automatic. The file type for each file will be assigned as PRG
or SEQ on the basis of the load address at the start of the file. The program also
checks for possible REL files. If a file may be part of a REL file, you will be asked if
this is possible. If you respond with a Y to this question, the program will attempt to
reconstruct the relative file.
Once the program has run to completion, all the files will have been recovered. The
first eight entries in the directory will be named FILE # 1, FILE # 2, and so on, and
will have to be renamed. However, this should be easy since you can load or read
the files.
264
NOTE:
In a few cases, such as the PRG files produced by WordPro™, PaperClip™ and Paint
Magic™, you may have to use the EDIT T&S program (Appendix E) or the MOD
ENTRY program (Section 13.8) to edit the file type byte to restore the correct file
type.
Problems on Sectors 2 to 18
If any one of these blocks cannot be read without error and is being used as part of the
directory file, you will be able to access some of the files on the diskette. However, any
files whose entries are in the damaged block or in subsequent blocks of the directory
file will not be accessible. Each block in the directory file can hold up to eight
directory entries.
The FIX DIRECTORY program from Appendix E can recover the files regardless of
the type of error on these sectors. To use the FIX DIRECTORY program to recover
from a read error on one of these sectors, simply load and run the program. When you
are asked which sector to fix, enter the sector number of the block to be recovered.
The program will attempt to read the sector you specified. If a read error is
encountered, you will be informed as to the type of error and presented with
three choices:
1. Null out the sector.
2. Bridge over the damaged sector.
3. Terminate the program.
You should always choose the null out the sector option at least once. If the sector
contains a 22, 23, or 24, READ ERROR, this should repair the damaged sector. If the
sector contains a 20, 21, or 27, READ ERROR, this attempt at repair may not work.
If it doesn't, choose the bridge over the damaged sector option. This causes the directory's
file chain to be altered to bridge over the damaged sector. If you choose this option,
your files will not be recovered immediately. The program will alter the directory file
chain, trace it and tell you which sector number to specify when you rerun the
program. To recover the lost files simply rerun the program and specify the sector
number that was indicated.
Once the sector can be read without error, it is checked to make sure it is an empty
sector. If any characters, other than nulls, are found, you will be asked if it is OK to
continue. This check is made simply so you have a chance to abort things in case you
specified the wrong sector number.
After this, everything is automatic. The file type for each file will be assigned as PRG
or SEQ on the basis of the load address at the start of the file. The program also
checks for possible REL files. If a file may be part of a REL file, you will be asked if
this is possible. If you respond with a Y to this question, the program will attempt to
reconstruct the relative file.
Once the program has run to completion, all the files will have been recovered. The
recovered files will have to be renamed. However, this should be easy now that you
can load or read the files.
265
NOTE:
In a few cases, such as the PRG files produced by WordPro™, PaperClip™ and Paint
Magic™, you may have to use the EDIT T&S program (Appendix E) or the MOD
ENTRY program (Section 13.8) to edit the file type byte to restore the correct file
type.
11.6 Recovering from a Full NEW
If you are reading this section in a panic in hopes of gaining absolution for your past
sins, relax; it's too late. However, if you are interested in learning how to avert a
future catastrophy, read on.
When you want to prepare a brand new diskette for first use or erase everything from
a previously used diskette, you use a full NEW disk command. A full NEW command
looks like this:
SYNTAX:
NO:diskette name, id
EXAMPLES:
OPEN 15,8, 15, "NO:LOTTA DATA,LD"
PRINT#15, "N: NEW GAMES, G5"
>N:WORDPROCESSINS 6,W6
Note that a diskette ID is always specified in a full NEW. If the diskette ID is
omitted, the disk drive performs a short NEW and just rewrites sectors 0 and 1 on
track 18.
If you are busily making a backup copy of your favorite diskette and you realize you
are standing there holding a blank diskette while your disk drive is merrily destroying
your master copy, don't panic. Reach over, pop the door of your drive open immed-
iately, and rescue your master diskette. At this point, you really don't care what the
manual says about not opening the door while the red drive-active LED is on. You
want that diskette now! A full track of data is being destroyed every two seconds
starting at track 1.
Once you have rescued the diskette, it is time to assess the damage. Try to load and
list the directory. If this works, use the CONFIRM ALL FILES program from
Appendix E to check for any problems. If you were quick, you may be pleasantly
surprised to find that all your files are safe. Copy all the intact files onto another
diskette and do a full NEW on the partially formatted diskette.
The reason it is possible to recover from a full NEW is related to how space is utilized
on the diskette. As you begin to store files on an empty diskette they are stored as close
as possible to track 18 (the directory track). The first file is stored beginning on track
17. Once track 17 is full, the next file is stored beginning on track 19. As more and
more files are added, the diskette gradually fills up. Tracks 1 and 35 are the last
266
tracks to be filled. This means that a partially filled diskette may have nothing stored
on tracks 1 to 10 and tracks 25 to 35. Since formatting begins on track 1 and progresses
upwards, none of the files would be erased from such a diskette until formatting
reached track 11. As long as you managed to pop the door before formatting reached
your files, everything would be safe. The only problem is that tracks 1 to 10 would be
formatted with a different embedded ID. Once the files were copied to another diskette,
this two-ID diskette should be reformatted with a full NEW.
11.7 Recovering a Damaged File
If one of the sectors used to store a file contains a read error, you will be unable to read
the entire file. You can use the RECOVER UNCLOSED program from Section 11.3
to recover the file as far as the damaged block. Trying to reclaim the rest of the file is
a bigger problem. Here are some suggestions.
The first thing to do is to trace the file. You can use the CONFIRM ALL FILES
program from Appendix E to print out the file chain for you. You want to specify the
printer output, trace file chains and print SUPER BAM options. Once you have these
printouts in hand, it is time for a bit of detective work.
The printout of the file chain for the program will list all the blocks in the file up to,
and including, the block containing the error. You want to make note of the track and
sector numbers of the last block before the error. Write these numbers down, you'll
need them in a minute.
Now look at the printout of the SUPER BAM. You should see several allocated blocks
that are not part of any file. These blocks will be highlighted in reverse field in the
printout. Your task is to determine which of these is the block that follows the
damaged block in the file chain. If you study the pattern of the numbers in the file
chain you should be able to make a pretty good guess. Let's look at an example.
Existing file chain:
17,00 17,10 17,20 17,08 17,18 17,06
Damaged block:
17,16
Allocated blocks not part of any file (from SUPER BAM):
17,01 17,02 17,04 17,12 17,14
The sector sequence seems to alternate between a low number and a high number
with a difference of 10 between them (0-10-20; 8-18; 6-16). Since the damaged block
has a high sector number, we would expect the next block to have a low sector
number. If we study the low numbers, we find that they seem to be decreasing by two
each time (10 -8 -6). As a result, we would expect the next sector in this file chain to be
track 17, sector 4.
Once you have determined the most likely next sector in the file chain, you can check
it using the TRACE CHAIN program listed below. Given a starting track and sector,
267
this program will trace the file chain to completion and display the file chain on the
screen. You can try several of the allocated blocks that are not part of any file as
starting points to see which produces the longest file chain.
PROGRAM: TRACE CHAIN
********************
0 REM 1541 USER'S GUIDE SECTION 11.7
1 REM COPYRIGHT: G. NEUFELD, 1984
2 :
3 REM TRACE & DISPLAY FILE CHAIN
4 REM INCLUDING THE SIDE SECTOR FILE
5 :
100 REM TRACE FILE CHAIN
HO PRINT" CCLR>TRACE FILE CHAIN"
120 INPUT" {DOWNJSTART AT (TRACK, SECTOR) "
»T,S
130 Z*=CHR*<0)
140 OPEN 15,8, 15, "10"
ISO GOSUB 370
160 OPEN 1,8,5, "#"
170 GOSUB 370
ISO PRINT" {DOWN> FILE CHAINCDOWN> "
190 NX=0
200 NX=NX+1
210 IF NX=7 THEN PRINT: NX=1
220 FC*=RIGHT*<STR*(T> ,2>+", "+RIGHT*("0"
+MID*(STR*<S>,2),2)+" "
230 PRINTFC*?
240 PRINT#15, "Ul:5 0"T;S
250 GOSUB 370
260 PRINT#15, "B-P:5 O"
270 GOSUB 370
280 GET#1,A*:T=ASC(A*+Z*)
290 GET#1,A*:S=ASC(A*+Z*>
300 IF T=0 AND S>0 THEN PRINT: PRINT" CDOW
N>END OF CHAIN": GOTO 420
310 IF T=0 AND S=0 THEN PRINT: PRINT" <DOW
N> VIRGIN SECTOR"T;S:GOTO 420
320 IF T=75 AND S=l THEN PRINT: PRINT" CDO
WN J VIRGIN SECTOR"T;S:GOTO 420
330 NS=20+2*(T>17)+(T>24)+(T>30)
340 IF T>35 OR S>NS THEN PRINT: PRINT" CDO
WNJBAD LINK"T;S:GOTO 420
350 GOTO 200
360 :
370 INPUT#15,EN,EM*,ET,ES
380 IF EN<20 THEN RETURN
390 PRINT:PRINT"<:DOWN>DISK ERROR"
400 printen;em*;et;es
410 :
420 CLOSE i: CLOSE 15: END
When you are certain that you know the track and sector numbers of the block that
follows the damaged one, you are ready to alter the file chain. You will use the EDIT
T&S program from Appendix E to do this. Here's the procedure:
Step 1: Load and run the EDIT T&S program.
Step 2: Insert the damaged diskette and ask to edit the last block before the damaged
one.
Step 3: Press PI to switch to hex edit mode.
Step 4: Edit the first two bytes on the top line of the display. These two bytes currently
are the track and sector numbers of the damaged block. Change them so that they are
the track and sector numbers of the block immediately following the damaged block
in the file chain. Remember, the numbers are in hexadecimal notation. Check the
table in Appendix B if you need help converting from decimal to hex.
Step 5: Press F6 to write out the revised sector to diskette.
Step 6: Press F5 to exit from the program.
You have now bridged over the damaged sector. One block of your file will be
missing, but now you can get at the rest of the file. Hopefully, you will be able to
reconstruct the missing data or program lines.
HINT:
If the file you have just recovered is a program file, it may cause your computer to
hang if you load it and try to edit it. However, you should be able to list it. If this
happens to you, read over Section 7.7 carefully. You should be able to create an ASCII
text file of your program. If you can, do so. Then, turn your computer OFF and ON,
type in the little merge routine from Section 7.7, and merge the ASCII copy of your
program back into memory. You should now be able to edit your program normally.
Naturally, with one whole block of lines missing, it won't run properly until you
reconstruct those missing lines.
11.8 Recovering from a Bad Replacement
Despite what you may have read, the save with replacement (SAVE "@0:name",8)
and the open for replacement (OPEN l,8,5,"@0:name,type,W") commands usually
work fine. As long as no disk errors are encountered during the writing of the
replacement file, everything works perfectly. However, if any disk error occurs and
the file is closed (this happens automatically With the SAVE with replacement
command), the new file will be incomplete and the old file appears to be lost.
269
Actually the old file is not really lost, it can be recovered, as we will see shortly. It is
still recorded on the diskette like a scratched file, but this time there is no cor-
responding directory entry to assist us in recovering the file. The old directory entry
has been modified so that the track and sector link now points to the replacement file.
This means that we cannot use our UNSCRATCHER program to recover it. We need
a program that creates a new directory entry and finds the correct track and sector
link for us.
The program, RECOVER FILE, listed below will recover any program, sequential
or user file stored on a diskette. It recovers scratched files as well as those lost by an
aborted replacement command. The only restriction is that the blocks used to store
the file must not be allocated in the BAM. You may have to VALIDATE the diskette
to free the blocks used to store the file.
The RECOVER FILE program is not as elegant as the FIX DIRECTORY program,
but it uses the same basic technique. Since this technique has never been documented
before, let me explain it. Line 230 of the program opens the specified file. When the
1541 receives this command, it checks the BAM for the first block available for use on
the diskette, and creates a new directory entry with a track and sector link that points
to this block. Since this is the same process it went through when the file to be
recovered was first opened, the new track and sector link inevitably points to the first
block in the file to be recovered! Lines 260-300 dig out the directory sector containing
the new entry and its location in that sector. (You could read through the directory
file to find the new entry but this is much more elegant.) In line 330 we issue an
initialize command. This automatically aborts (not closes) any active disk files. (We
don't want to close the file normally because this would overwrite the first block in the
file we want to recover.) We have now created a new directory entry that points to the
file we want to recover. However, it has the wrong file type; it is an unclosed file.
Lines 370-430 read in the directory sector containing the entry and correct the file
type byte. The only job left is to validate the diskette to correct the BAM. Line 460
does this.
The resurrected file will show up in a directory listing as having a length of zero
blocks. This is a cosmetic problem. The entire file has been recovered and all blocks
have been allocated in the BAM and are protected. If you copy the file onto another
diskette, the new file will show the correct length. However, if you don't want to copy
the file onto another diskette and this cosmetic problem bothers you, here's how to
solve it First, use the CONFIRM ALL FILES program (Appendix E) to find out the
file's actual length. Then use the MOD ENTRY program (Section 13.8) to change the
directory entry so that it indicates the proper length.
****************
PROGRAM: RECOVER
****************
NOTES:
1. The RECOVER FILE program recovers any stray files on the diskette. If your
diskette contains any scratched files, you may have to run the program several
times until you recover the file you want
270
2. Be sure to use a new file name for the recovered file. Otherwise you will get a 63,
FILE EXISTS, error and the program will abort
0 REM 1541 USER'S GUIDE SECTION 11.8
1 REM C0PYRI6HT: G. NEUFELD, 1984
2 :
3 REM RECOVER FILE WITHOUT DIRECTORY EN
TRY
4 :
lOO REM RECOVER FILE FOR C-64/V-20 & 154
1
HO QT*=CHR*<34)
120 PRINT"CCLR3 RECOVER FILE"
130 INPUT" {DOWN> NAME FOR RECOVERED FILE"
;rn*
140 input-file type (p/s/u)";rt*
150 if rt*="s" then ft= 129: goto 200
160 if rt*="p" then ft=130:g0t0 200
170 if rt*="u" then ft=131:goto 200
180 print" cdown>enter only *p', *s' , or
'U'":GOTO 140
190 :
200 DEF FN A <X)=PEEK<57) +256*PEEK<58)
210 OPEN 15,8, 15, "10"
220 LN=FN A(X):GOSUB 520
230 OPEN 1,8,5, "0:"+UN*+","+UT*+",W"
240 LN=FN A(X):GOSUB 520
250 :
260 REM DIG OUT POINTER TO DIRECTORY ENT
RY
270 PRINT#15, "M-R"CHR* < 144) CHR* (2) CHR* ( 1
)
280 GET# 1 5 , A* : SE= ASC < A*+CHR* ( O ) )
290 PRINT#15, "M-R"CHR* ( 148) CHR* <2> CHR* < 1
)
300 GET# 1 5 , A* : PT= ASC ( A*+CHR* ( O ) )
310 :
320 REM INITIALIZE TO CREATE UNCLOSED FI
LE
330 PRINT#15,"I0"
340 CLOSE 1
350 :
360 REM CHANGE ENTRY TO CORRECT FILE TYP
E
370 OPEN 2,8,6, "#"
380 PRINT#15, "Ul:6 O 18"SE
390 LN=FN A(X):GOSUB 520
400 PR I NT# 1 5 , " B-P : 6 " PT
410 PRINT#2,CHR*(FT);
420 PRINT#15, "U2:6 O 18"SE
271
430 LN=FN A(X):G0SUB 520
440 :
450 REM VALIDATE DISKETTE TO ALLOCATE BL
OCRS
460 PRINT#15„ "VO"
470 INPUT#15,EN,EM*,ET,ES
480 IF EN<20 THEN PRINT" CDOWN> FILE "QT*;
RN*;QT*" HAS BEEN RECOVERED": GOTO 560
490 PRINT" CDOWNJWARNING: BAM IS NOT CORR
ECT"
500 PRINT" {DOWN J FILE RECOVERED BUT NOT P
ROTECTED'TGOTO 560
510 :
520 INPUT#15, EN, EM*, ET, ESzPRINTLN
530 IF EN<20 THEN RETURN
540 PRINT" <DOWN>DISK ERROR AT LINE"LN
550 printen;em*?et;es
560 CLOSE 2: CLOSE 15
570 END
11.9 Recovering a Physically Damaged Diskette
Despite all the warnings you have heard and read, you probably still like to sip a cup
of coffee or a soft drink while using your computer. Eventually the inevitable happens.
You end up with a very wet and soggy diskette. Of course, this diskette is the one with
all your current work on it and there is no backup. Why else would it be lying around
so conveniently?
Well, now that the inevitable has happened, what do you do about it? You treat it
somewhat like a kid that just fell in a puddle. You take off its wet clothes, wash it up a
bit, and put it in dry clothes.
The first step is to remove the mylar disk from its protective jacket. Use a sharp knife
to pry up the flaps that hold the two sides of the jacket together and open it up.
Remove the mylar disk carefully (try to remember which side is up). Do not touch the
recording surface with your skin as you do this (a pair of gloves makes this easier).
Now rinse off the disk with clean water. If you have hard water, a little wetting agent
(like photographers' PHOTO FLO) will help prevent streaking and mineral deposits.
Set the disk on a soft, lint free surface to dry.
Now for the dry clothes. You can sacrifice a new diskette to get a clean, dry jacket.
Or, if you have a floppy disk type head cleaner (see Section 12.2), you can use its
permanent jacket as a temporary home for your freshly cleaned diskette. Insert the
diskette into its new jacket. Try loading a directory. You may be pleasantly surprised
to find that everything is back to normal. If it is, how about a backup, or maybe two?
If you can't load the directory, you may have put the diskette in the jacket upside
down. Try putting the diskette in the drive upside down. Still no luck? Too bad!
272
11.10 Conclusion
This chapter has described how to recover from most types of diskette problems. As
you may have noticed, recovery is usually possible but can be difficult. The disk
utility programs listed in this chapter and in Appendix E are very powerful tools and
represent the current state of the art in file recovery. They allow automated recovery
of files from almost any kind of problem, even from problems that eminent authorities
have claimed make the files completely unrecoverable! However, any powerful tool
can be misused. You should always practice using a program with a test diskette
before you operate on one that is important or valuable.
Despite these powerful utilities, it is always easier to dig out your archival copy of a
program or data file than to recover a damaged diskette. Be safe, not sorry! Backup
your diskettes regularly! If you don't have a backup program, you can use the
BACKUP program from Appendix E. It is not the fastest backup program available
(it takes about 25 minutes to backup a diskette), but it does a reasonable job and does
not abort if it encounters a read or write error.
273
CHAPTER
12
CARE AND MAINTENANCE
This chapter describes how to care for your 1541 disk drive and keep it in top shape.
Included are such topics as cleaning the record/play head, permanently changing the
device number and preventing drive wheel slippage. This chapter also includes a
listing of a disk utility program that allows you to align your 1541 disk drive without
using an oscilloscope or other expensive equipment.
12.1 General Operating Hints
Many of the problems that people encounter with their disk drives are the result of a
poor operating environment. Dust, smoke particles and stray magnetic fields all
cause problems. Keep your working environment clean and smoke free. Be sure to
keep your diskettes away from sources of magnetic fields such as TV sets (this includes
your monitor) and electric motors. These simple steps will eliminate many potential
errors.
Another source of problems is switching your disk drive ON or OFF with a diskette
in it. The resulting sudden surge of current through the record/play head may produce
a magnetic spike on the diskette surface. If the spike occurs in the wrong place, you
will make one block of information on the diskette unreadable. Since the record/play
head spends much of its time on track 18, you are likely to destroy the diskette's
directory. This is the most common cause of a 23, READ ERROR.
Heat buildup is also a cause of problems with the 1541 drive. The case is not well
ventilated. As a result, the temperature inside the case gradually rises, over a 30
minute period, until it reaches a stable value. At this point, the temperature inside
the case may be high enough to exceed the maximum design operating temperature
of some of the integrated circuit chips. This can cause a chip to fail. The high temper-
ature can also cause alignment problems as indicated in Section 12.4. To minimize
the temperature problem follow these operating rules:
1. Always place the drive on a hard, clean surface. If it is placed on a soft surface, the
bottom air vents may be restricted or blocked.
2. Turn off your 1541 when it is not in use.
3. Do not place papers, diskette envelopes or diskettes on top of the drive. You may
restrict or block the upper vents.
275
If you have a severe heat problem, you may have to install a fan or cut larger vents in
the case.
12.2 Cleaning the Record/Play Head
There are a lot of conflicting opinions about how often you should clean a disk drive's
record/play head. Some experts claim you should clean it once a month while others
suggest less frequent cleaning. I have worked with disk drives for several years, and
tend to favor yearly cleaning. One disk drive that I use regularly has not been cleaned
for three years and there is no noticeable buildup of residue on the head. However, I
always use medium grade diskettes. If you use inexpensive off-brands, you may get a
more rapid buildup.
The easiest way to clean the record/play head is to use a commercial head cleaner.
They look somewhat like a floppy diskette but instead of having a plastic disk inside
the jacket, they have a slightly abrasive disk impregnated with a cleaning solvent. To
use one of these cleaners simply insert the cleaning diskette in the drive and attempt
to load a directory several times. This causes the drive motor to switch on. The
rotating disk cleans the head.
You can also clean the head using a cotton swab moistened with denatured alcohol
(not rubbing alcohol!). However, you must disassemble the 1541 to gain suitable
access to the record/play head, so this procedure is not recommended for the novice.
12.3 Permanently Changing the Device Number
The procedure outlined below is useful when you own two or more disk drives. By
permanently altering the device numbers you avoid the bother of changing the device
numbers every time you turn on your system. The procedure involves using a knife to
cut one or two jumpers on the printed circuit board inside the disk drive. You will
need a small Phillips (+) screwdriver and a sharp knife. If you feel uncomfortable
about tinkering around with things electronic, have your dealer do it for you.
Note that there are two types of 1541 disk drives: the long board and the short board
versions. The location of the jumpers mentioned in Step 5 is different for the two
versions, so you will have to know which version you have. In most cases you can tell
by the color of the exterior case. Most, if not all, of the original, long board versions
have ivory colored cases. If your drive has a brown case, it is probably the newer,
short board version. Once you open the case you will be able to tell for sure. The
printed circuit board on a long board version is about 12.5 inches (31 cm) long. It is
only about 9.5 inches (24 cm) long on the short board version.
1. Remove all connecting cables, including the power cable, from your drive.
2. Turn your disk drive upside down on a flat, level, stable surface and remove the
four small Phillips screws that hold the case together. They are recessed into the
bottom — two at the front and two at the rear of the drive.
3. Carefully turn the drive upright and remove the top cover.
276
4. There may be a metal plate that covers the printed circuit board. This must be
removed. It is held in place by two small screws. Both are on the left-hand side as
you face the front of the drive. If you have a long board version, be careful of the
two wires that lead to the green power light at the front of the drive. These wires
are attached to a plug that plugs into the circuit board. You can unplug them but
make a diagram showing how and where they are attached.
5. Locate the device number jumpers. They look like two semicircles joined by a
narrow strip. The diagram below indicates the location of the jumpers on the two
versions of the 1541.
• •
Jumpers
IS
Dps
•
0P8
t
»
•
•
•
D
pi
>
»
flw
Jumpers
•
© 2
<D 1
•
Short Board
Top View of Printed Circuit Boards 1541 Drive
Long Board
On the long board version the two jumpers are labeled one and two. They are not
labeled on the short board version. On the short board version jumper one is the one
closest to the front of the drive near the transistor marked Q3. Jumper two is the one
near the capacitor marked C46.
6. Use a knife to cut through the narrow trace connecting the two semicircles for one
or both of the jumpers. Use the following table to determine which jumpers to cut.
Device #
Jumper 1
Jumper 2
#8
#9
#10
intact cut intact
intact intact cut
#11
cut
cut
EXAMPLE: For a 1541 drive to
be device# 9, cut jumper one
and leave jumper two intact.
7. Replace the metal cover and fasten it in place with the two small screws. If you
have a long board version, be sure to reconnect the power light wires.
277
8. Replace the top of the case and fasten it in place with the four screws. Now
reconnect the power cord and cables.
Since this procedure is recommended in your 1541 User's Manual (page 40), it should
not affect your warranty. To convert a drive back to device# 8, bridge the jumper you
cut with solder.
12.4 The Alignment Problem
The most common problem with the 1541 disk drive is head alignment. The
mechanism that positions the record/play head tends to shift out of alignment easily.
When this happens, the head will not be positioned at the correct distance from the
center of the rotating disk to read the tracks and sectors correctly. A greatly enlarged
view of the disk and head might look something like this:
Track 16
Track 17
Head should
be positioned
accurately on
the track.
Correct Misaligned
If your drive's head is not properly aligned, read errors will occur frequently. Watch
the red drive-active LED on your drive while a long program is loading. If it blinks
off and on occasionally, your drive is having trouble reading the diskette.
An alignment problem may develop gradually or suddenly. Most alignment problems
are caused by the slipping of the drive wheel on the shaft of the stepper motor which
is used to move the record/play head from track to track. Slippage occurs because the
drive wheel is only a press fit on the stepper motor shaft. There is no set screw or
keyway to maintain the correct alignment. When the 1541 has been operating for a
long time, it gets hot inside, the metal parts expand and the drive wheel may slip if
stress is applied. Normal accessing of files does not produce enough stress to cause
problems. However, slippage will occur if there is a bump to track 1. When the drive
does a bump, the stepper motor attempts to move the head outward a total of 45
tracks. This is done to ensure that the head is positioned over track 1. A lobe on the
stepper motor drive wheel hits a mechanical stop to prevent the head from moving
past track 1. It is the hammering of this lobe against the mechanical stop that causes
the clatter each time you format a diskette or when the drive is attempting to recover
from an error. If a bump occurs when your drive is hot, some slippage is likely.
You can minimize the possibility of slippage by following these general operating
rules:
1. Turn off your 1541 disk drive when it is not in use. This minimizes the heating
problem.
2. Do not format a diskette when your disk drive is hot.
278
3. Do not load any DOS protected diskette that causes the drive to err out when your
disk drive is warm. Many disk drives have been thrown out of alignment by the
use of such programs. A DOS protection scheme does not have to hammer your
drive to death to be effective. You should return as defective any commercial
diskette that causes a bump each time the program loads. Don't take chances. It's
your drive that will be damaged!
If you follow these rules, you will minimize, but not eliminate, the alignment problem.
When your drive finally does go out of alignment, you can realign it yourself using the
procedure described in Section 12.6.
12.5 Checking the Alignment of your 1541
Normally you need an oscilloscope and a special alignment diskette to align a disk
drive. However, the program listed below can be used to check the alignment of your
drive and even realign it (see Section 12.6). All you need is your Commodore 64
computer, your 1541 drive, a copy of the ALIGN 1541 program and a standard
diskette. You can use any formatted diskette as your standard. However, well use
your 1541TEST/DEMO diskette in our examples.
Here's what to do:
1. Load and run the ALIGN 1541 program. If your drive is very badly out of
alignment, you may have to load the program from cassette or borrow a friend's
drive.
2. Insert your 1541TEST/DEMO or other standard diskette.
If your drive is reasonably well aligned, you should see a display like the one shown
below. If your drive is not aligned, the track numbers and head position message may
be different.
ALIGN THE 1541 DRIVE
TRACK # FROM STEPPER: 18
HEAD POSITION: CLEAN READ OF TRACK
TRACK # AS READ: 18
SECTOR # AS READ: 4
ID OF TRACK READ: ZX
COMMANDS:
Fl = MOVE HEAD OUT (LOWER TRACK #)
F3 - MOVE HEAD IN (HIGHER TRACK #)
F7 = TERMINATE PROGRAM
I - INITIALIZE (TO TRACK 18)
279
Let's analyze this display line by line.
TRACK # FROM STEPPER: 18
This tells us that the stepper motor has positioned the 1541's read/write head to the
position where it expects to find track 18. As the diskette rotates past the record/play
head a special machine language routine (that the ALIGN 1541 program stored in
the disk drive's RAM memory) attempts to read the header blocks of the sectors as
they move past.
HEAD POSITION: CLEAN READ OF TRACK
This line of the display tells us whether our routine was successful in reading and
interpreting a header. There are three possible messages that can be displayed:
CLEAN READ OF TRACK — indicates that a header was read successfully.
CLOSE TO HALFWAY POINT — found a header but it was junk (high tone).
EXACTLY HALFWAY BETWEEN — couldn't even find a sync mark (low tone).
At this point you should be getting the CLEAN READ OF TRACK message. You
can force a display of the third message by opening the drive door and removing the
diskette. This will not harm your drive or diskette. You can even switch diskettes and
try a different standard. The program doesn't mind!
TRACK # AS READ: 18
SECTOR # AS READ: 4
ID OF TRACK READ: ZX
These three lines of the display tell us the contents of the header if we managed to
read one successfully. If you watch the display, you will see the sector number change
as the sectors flip past. Since we are only reading a header about once a second, we
don't get to see the sectors in sequence.
COMMANDS:
Fl - MOVE HEAD OUT (LOWER TRACK #)
F3 - MOVE HEAD IN (HIGHER TRACK #>
F7 - TERMINATE PROGRAM
I ■ INITIALIZE (TO TRACK 18)
The last part of the display shows us our options. Each time you press Fl, the head
moves outward half a track. Yes, you read that correctly — half a track. Try it and
see what happens. The line that gives the track number from the stepper motor
should now indicate that you are on track 17.5! Depending on the alignment of your
drive, the track number as read may be either 17, 18 or ??. If you are getting ?? as the
track number, you should be hearing either a low or a high-pitched chime and have
either a CLOSE TO HALFWAY POINT or EXACTLY HALFWAY BETWEEN
message on the screen.
Now that you know what the screen display means, practice moving the record/play
head around. Remember, you press:
280
Fl to move half a track outwards (towards track 1).
F3 to move half a track inwards (towards track 35).
I to return to track 18.
F7 to terminate the program.
If your drive is reasonably well aligned, you should be able to read all the tracks
cleanly. On the halftracks you may read the upper track, the lower track, or something
in between.
Now that you know how to move the head around, let's take a look at the amount of
hysteresis (his-tur-e-sis) in the head positioning mechanism. This is a fancy way of
talking about the amount of play there is in the mechanism. Here's what to do:
1. Use F3 to move the head to track 19 (stepper motor).
2. Use Fl to move the head out to track 17.5.
3. Write down the information about what is being read.
4. Use Fl to move the head out to track 16.
5. Use F3 to move the head in to track 17.5.
6. Write down the information about what is being read.
The difference between what you found in Steps 3 and 6 gives you an indication of the
amount of play in the head-positioning mechanism. On an optimally aligned drive you
should get the message, CLOSE TO HALFWAY POINT whether you approach
track 17.5 from a lower or higher track number. If you always get a large discrepancy
(a clean read of 17 from one direction and a clean read of 18 from the other) with
several different standard diskettes, there is an excessive amount of play or friction in
the head-positioning mechanism. You should take your drive to a competent technician
or at least lubricate the rods that the head carriage slides on (see Step 11 in Section
12.7).
The reason we chose track 17.5 as the place to make our hysteresis check was that
1541 drives are normally aligned using track 17. However, there is nothing partic-
ularly sacred about this track. If you decide to use some other point as your standard
reference point, make sure it is near the middle of the recording area of the diskette.
By now you should have a pretty good idea whether your drive should be realigned. It
should be realigned if you found:
1. That any track could not be read cleanly (i.e., you received either the CLOSE TO
HALFWAY POINT or EXACTLY HALFWAY BETWEEN message when the
stepper motor track number was a whole number).
2. That there were any tracks where the track number, as read, did not agree with
the track number from the stepper motor.
You don't really have to worry about what happens between tracks unless you are in
the process of realigning your drive. In that case you would like to adjust the drive so
you get the message, CLOSE TO HALFWAY POINT, whether you approach track
17.5 from a lower or higher track number.
281
*******************
program: align 1541
*******************
0 REM 1541 USER'S GUIDE SECTION 12.5
1 REM COPYRIGHT: G. NEUFELD, 19S4
2 :
3 REM CHECK ALIGNMENT AND/OR RE-ALIGN A
1541
4 :
ioo printmcclr><:downj 1541 ALIGNMENT"
HO PRINT" CDOWN 2> INSERT STANDARD DISK"
120 PRINT" {DOWN 2>PRESS {RVS J RETURN CROFF
> WHEN READY"
130 :
140 REM MACHINE CODE ROUTINE TO READ A
HEADER
ISO REM RESIDES AT *0300 (BUFFER #0)
160 :
170 DATA 169,48: :REM LDA #*30
180 DATA 133,69: :REM ST A *45
190 DATA 169, OO: :REM LDA #*00
200 DATA 133,63: :REM STA *3F
210 DATA 76,177,243 :REM JMP *F3B1
220 :
230 D* (O) ="00" : D* < 1 ) = "01 " : D* (2) =" 10" : D* (
3) ="11"
240 SID = 54272
250 DIM FD*(16)
260 FD*<0)="
270 FD*<1)="CLEAN READ OF TRACK
280 FD*(2)="CL0SE TO HALFWAY POINT
290 FD*(3>="EXACTLY HALFWAY BETWEEN "
300 FD*(9)="CL0SE TO HALFWAY POINT
310 T=18: Nl*="?" : N2*="?" : TR=255
320 GET A*: IF A*OCHR*(13> GOTO 320
330 :
340 OPEN 15,8, 15, "I"
350 PRINT" CCLR>"
360 :
370 REM READ THE DISK CONTROLLER PORT
380 :
390 PR I NT# 15," M-R " CHR* ( O ) CHR* ( 28 )
400 GET#15,A*:IF A*=""THEN A*=CHR*(0>
410 A=ASC(A*>
420 CV=3 AND A
430 A= ( 159ANDA) OR (96+32* < <T>17) + (T>24) + (
T>30> ) )
440 PRINT#15, "M-W"CHR* <0> CHR* (28) CHR* ( 1 >
CHR* (A OR 4)
450 :
282
460 REM DISPLAY VALUES
470 :
480 PRINT" {HOME* <DOWN} ALIBN THE 1541
DRIVE"
490 PRINT" CDOWN} TRACK # FROM STEPPER: "T"
{LEFT}
500 PRINT" {DOWN} HEAD POSITION: "FD*(E)
510 T*=STR* ( TR ) : S*=STR* < SE ) : I F E<> 1 THEN
T*="??" : ni*="?" : N2*='"?" : s*="??"
520 PRINT" {DOWN} TRACK # AS READ: "RIGHT
$(T*,2)
530 PR I NT "SECTOR # AS READ: "RIGHT* (S*, 2
>
540 PRINT" ID OF TRACK READ: "N1*;N2*
550 PRINT" {DOWN 2 } COMMANDS : "
560 PRINT" {DOWN} Fl = MOVE HEAD OUT (LO
WER TRACK #)
570 PRINT" F3 = MOVE HEAD IN (HIGHER TR
ACK #>
580 PRINT" F7 = TERMINATE PROGRAM"
590 PRINT" I = INITIALIZE (TO TRACK 18
) "
600 GET A*:A=ASC(A*+CHR*CO) )
610 IF A=136 GOTO 800
620 IF A=133 AND T>1 THEN C=-l:GOTO 690
630 IF A=134 AND T<35 THEN C=l:GOTO 690
640 IF A=73 THEN PRINT#15, "I":T=18:E=0: A
=214:G0T0420
650 GOTO 890
660 :
670 REM MOVE HEAD ONE HALF-TRACK IN OR O
UT
680 :
690 CV=(CV + OAND3
700 T=T+C*.5:IFT<1 THENT=1
710 IFT>36THENT=36
720 B=A AND 252
730 C=B+CV
740 PRINT#15, "M-W"CHR* (O) CHR* (28) CHR* ( 1 )
CHR*(C)
750 E=0
760 GOTO 390
770 :
780 REM TERMINATE PROGRAM (DRIVE OFF)
790 :
800 PRINT#15, "M-W"CHR$ CO) CHR* (28) CHR* ( 1 )
CHR* (240)
810 FOR I=SID TO SID+23:P0KE 1,0: NEXT
820 FOR K=1T010: GET A*: NEXT
830 CLOSE 15: END
840 :
283
850 REM ATTEMPT TO READ ANY HEADER
860 :
870 REM READ & SEND MACHINE CODE ROUTINE
880 :
890 RESTORE :C*=""
900 FOR K=l TO ll:READ X : C*=C*+CHR* ( X > : N
EXT
910 PRINT#15, "M-W"CHR* (O) CHR* (3) CHR* (11)
C*
920 :
930 REM PUT JMP JOB IN THE JOB QUEUE
940 :
950 PRINT#15, ,,M-W"CHR*<0)CHR*<0) CHR* (DC
HR*(208>
960 :
970 REM WAIT FOR JOB TO FINISH
980 :
990 PRINT#15, "M-R"CHR*(0)CHR* (O)
1 OOO GET# 1 5 , E* : E= ASC ( E*+CHR* ( O ) )
1010 IF E>127 SOTO 680
1020 :
1030 REM "E" IS FDC ERROR CODE RETURNED
1040 IF E=3 THEN HZ=1000:G0SUB 1190
1050 IF E=2 OR E=9 THEN HZ=1500:G0SUB 11
90
1060 IF EOl GOTO 390
1070 :
1080 REM CLEAN READ SO DIG OUT ID, TRAK
& SECT
1090 :
llOO PRINT#15, "M-R"CHR*(22)CHR*(0>CHR*(4
)
1110 GET#15„N1*
1120 GET#15,N2*
1130 GET#15,X*:TR=ASC(X*+CHR*(0> )
1 140 GET#15, X*: SE=ASC (X*+CHR* (O) )
1150 GOTO 390
1160 :
1170 REM BLEEP IF NO SYNC
1180 :
1190 FOR I=SID TO SID+23:POKE 1,0: NEXT
1200 POKE SID+5,9
1210 POKE SID+6,9
1220 POKE SID+24, 15
1230 POKE SID+l,E+20
1240 POKE SID, 177
1250 POKE SID+4, 17:FOR I=1T050:NEXT
1260 REM POKE SID+24, O
1270 RETURN
284
1280 POKE Ai, 85: POKE S1„150:P0KE Wl,65
1290 REM FOR K=l TO lOOOzNEXT
1300 RETURN
12.6 Realigning Your 1541
When you have your 1541 professionally realigned, the service technician uses an
oscilloscope, an alignment program and a special alignment diskette. You probably
don't have access to equipment like this. However, that doesn't mean you can't realign
your drive. The utility program listed in Section 12.5 makes it possible for you to
check the alignment of your 1541 and realign it without the use of any expensive
equipment.
Know what you are doing! Mistakes can be costly.
Don't attempt to realign your drive unless you have practiced using the ALIGN 1541
program to check your drive's alignment. Read the instructions below and those
given in Section 12.5 several times before you start on this procedure. If you are a
novice in things electronic, you may not want to realign the drive yourself.
It should take you less than an hour to disassemble your drive, adjust the alignment
and reassemble everything. To disassemble your drive and realign it you will need
the following tools:
1. #1 and #2 Phillips (+) screwdrivers.
2. A previously formatted diskette to use as your standard.
3. A copy of the ALIGN 1541 utility program.
To adjust the drive speed you will also need:
4. A fluorescent light.
5. A very small, flat (-) jeweler's screwdriver.
Here's what to do:
1. Load and run the ALIGN 1541 program listed in Section 12.5. Check the screen
display. The head should be positioned on track 18 and should be reading the
headers for track 18 cleanly. If you are not getting the head-position message,
CLEAN READ OF TRACK, press the Fl key to move the head to track 17.5.
Once you can see which track is being read, compare it to the stepper motor track
position. If the difference between these is greater than one (e.g., you are reading
track 16 while positioned over track 18), the drive is so far out of alignment that
you should have it professionally realigned. It is beyond the normal adjustment
range.
2.
Once you are satisfied that your drive can be aligned, remove all connecting
cables, including the power cable, from your drive.
285
3. Turn your disk drive upside down on a flat, level, stable surface and use your #1
Phillips screwdriver to remove the four small screws that hold the case together.
They are recessed into the bottom, two at the front and two at the rear of the
drive.
4. Carefully turn the drive upright and remove the top cover.
5. Use your #1 Phillips screwdriver to remove the six small screws that hold the
drive and the bottom of the case together. They are down inside the edge of the
case at the bottom, three on each side of the drive.
6. Carefully lift the drive out of the bottom of the case and turn it upside down on a
piece of foam to the left of the case. Be careful that you don't tear off the wires
leading to the green power-on LED.
7. Use the diagram below to locate the two screws that hold the stepper motor, the
strobe disk and the speed adjustment access hole.
J
k
E
\ Motor /
3
Stepper \
.Strobe
/ J>\ Disk
Loosen
T^
Speed
Adjust
□
r
Bottom View Case Removed — Short and Long Board Versions 1541 Drive
8. Leaving the drive upside down, carefully reconnect the power cord and serial bus
cable and switch ON the drive.
9. With the drive still upside down, run the ALIGN 1541 program and insert your
standard diskette into the drive. Although you can use any formatted diskette as
a standard, you may want to use the 1541TEST/DEMO diskette that came with
your drive.
10. Use your #2 Phillips screwdriver to loosen the two screws holding the stepper
motor in place. The motor should not be loose but you should be able to turn it by
hand.
11. Look at the video display. It should indicate that the head is positioned above
track 18 and that the head is reading track 18 cleanly. If you don't get this result,
turn the stepper motor case slightly clockwise or counterclockwise until you do
get this display. If you are unable to adjust the motor to get this display, there has
been a great deal of slipping and you should let a professional realign your drive.
12. Use the F3 key to move the record/play head to track 19 and then use Fl to bring
it back out to track 17.5. Turn the stepper motor case clockwise or counter-
clockwise until the display indicates that you are near the midpoint between
tracks 17 and 18. A slight turn one way should allow you to read track 17 and a
slight turn the other way should give you track 18.
13. Use the Fl key to move the record/play head out to track 16 and then use F3 to
bring it back in to track 17.5. Turn the stepper motor case clockwise or counter-
clockwise slightly until the display indicates that you are near the midpoint
14. Since there is some play (hysteresis) in the head-positioning mechanism you will
have to repeat Steps 12 and 13 making smaller and smaller adjustments each
time until the head is about equally close to the midpoint between the two tracks
whether the head approaches from lower or higher track numbers. Remember
that on an optimally aligned drive you will be able to read all tracks cleanly and
should get the CLOSE TO HALFWAY POINT message (high tone) whether the
head approaches track 17.5 from higher or lower track numbers. Once you are
satisfied with the head position, carefully tighten the adjustment screws.
15. Since the drive is apart, you may as well adjust the drive speed. The outer ring of
the strobe disk (marked 60) should appear stationary when viewed under
fluorescent light. Use the fine screwdriver to turn the small screw clockwise or
counterclockwise to adjust the drive speed. Be careful not to short out any traces
on the printed circuit board as you insert the screwdriver through the small
access hole.
16. Remove your standard diskette and turn OFF the drive. Disconnect the power
cord and serial bus cable. Reassemble the drive. Your realignment job is complete.
12.7 Preventing Drive Wheel Slippage
Since alignment problems on the 1541 are common and are usually the result of
slippage of the drive wheel on the shaft of the stepper motor, various ways have been
suggested to lock the shaft and drive wheel together. Epoxy or crazy glue by them-
selves do not solve the problem permanently! There must be a solid mechanical
connection if the fix is to last. The modification described below is fairly easy to do,
requires relatively simple tools, and locks the stepper motor shaft and the drive wheel
together permanently so there is no possibility of any slippage.
287
Know what you are doing! Mistakes can be costly.
The modification described below will definitely void any warranty on your disk
drive. If your disk drive is still under warranty, you should not modify it. Do not
attempt this modification if your drive is badly out of alignment! Make sure you read
all the instructions and understand them before you begin. This modification requires
careful work. If you are not comfortable with things mechanical, you should have
your dealer make the modification for you.
You will need the following tools to carry out this modification.
1. #1 and #2 Phillips (+) screwdrivers.
2. A Dremel™ tool or other hand grinder.
3. A thin (1/32 inch), metal cutting wheel for the grinder.
4. Some five-minute epoxy cement.
5. A steel dressmaker's pin.
6. Wire cutters.
Here's what to do:
1. Load and run the alignment program listed in Section 12.5. Follow the procedure
described in that section for checking the head alignment and hysteresis. If you
are not able to read the correct tracks cleanly, you should realign your drive
before proceeding any further. You do not want to permanently misalign your
drive!
2. Once you are satisfied that your drive is reasonably well aligned, remove all
connecting cables, including the power cable, from your drive.
3. Turn your disk drive upside down on a flat, level, stable surface and remove the
four small Phillips screws that hold the case together. They are recessed into the
bottom, two at the front and two at the rear of the drive.
4. Carefully turn the drive upright and remove the top cover.
5. There may be a metal plate that covers the printed circuit board. This must be
removed. It is held in place by two small screws. Both are on the left-hand side as
you face the front of the drive. If you have a long board version, be careful of the
two wires that lead to the green power light at the front of the drive. These wires
are attached to a plug that plugs into the circuit board. You can unplug them but
make a diagram showing how they are attached.
6. Using your #1 Phillips screwdriver, remove the seven screws that hold the printed
circuit board in place. There are five screws that go through the board itself.
There are two more screws through the right-hand side of the case that hold the
heat sink on.
288
7. Tip the printed circuit board to one side so you can get at the drive mechanism
beneath it. You may have to unplug one or more cables from the board to do this.
Be sure to make a sketch of the position of any plugs before you unplug them.
8. With the help of the diagram below, locate the stepper motor's drive wheel.
uzr
/ \
d
WMMmmffli
Top View with Printed Circuit Board
Removed — Short and Long Board
Versions 1541 Drive
9. Use the Dremel™ tool to cut a narrow slot across the top of the stepper shaft and
one of the lobes on the drive wheel as indicated above. The dressmaker's pin
should just fit into the slot when you are done. Take your time and use a very light
touch. Too much heat may cause the drive wheel to shift on the shaft. You may
want to mask off the work area to keep flecks of metal out of the drive mechanism.
When you are finished, use a vacuum to clean up the area.
10. Trim the dressmaker's pin to the correct length and epoxy it in place. Once the
glue is dry, partially reassemble the drive and check the alignment. You may
want to touch up the alignment and drive speed before final reassembly.
11. If you found a fair bit of play in the head-positioning mechanism when you checked
the alignment, you may want to lubricate the two round bars that the head-
positioning mechanism slides on. Use a very small amount of a high grade oil
(such as clock oil) or a lightweight grease (such as a lithium based bicycle grease)
on a cotton swab to lubricate them sparingly.
Once you have made this modification, you should realign your drive (see Section
12.6). This time it should stay in alignment!
289
CHAPTER
13
A BIT OF BACKGROUND
This chapter is designed to take some of the mystery out of the actual operation of
your 1541. You don't need to know about the inner workings of your car to be able to
drive it. However, if you run into trouble on the road, it helps to know a bit about
spark plugs, distributors and carburetors. Similarly, you do not need to know about
the inner workings of your disk drive to use it effectively. But, a bit of background
comes in handy when you start having problems. Topics discussed in this chapter
include the recording process, how the directory is organized, and how sequential,
program, user and relative files are stored.
13.1 Inside a Floppy Diskette
A floppy diskette consists of three parts: a plasticized outer protective jacket, an inner
liner and a circular disk of thin mylar with a magnetic coating.
inner liner
write protect notch
mylar disk
outer jacket
index hole
access slot for read/write head
mouse bites
291
Outer Protective Jacket
The outer protective jacket is made of a relatively stiff plasticized material. Its job is
to protect the fragile disk inside. There are several holes and notches in the outer
jacket. The large central hole allows the wedge-shaped central hub in the drive to
grip the inner disk and turn it. The large elongated slot allows the drive's read/write
head to contact the disk so it can read or write information. The write protect notch
can be used to protect information stored on a diskette. If the write protect notch is
covered with an opaque tab, the drive will not write (record) onto the diskette. The
mouse bites on the edge of the jacket near the access slot are strain relief notches to
prevent the plastic disk from binding. The index hole is not used on any Commodore
drives. It is used on some brands of drives to determine the position of the mylar disk.
If you rotate the inner disk carefully you will see one or more holes in the disk that
become visible as the disk is turned.
Inner Liner
The inner liner is a loosely woven, slippery material. Its main job is to reduce friction
between the jacket and the rotating disk. Since the material is fibrous, it also cleans
the surface of the disk as it turns. Small particles of dust, smoke and grit are collected
from the disk and become embedded among the fibers.
Mylar Disk
The inner disk is a circular piece of mylar. Better quality diskettes have a ring of
reinforcing material around the large central hole. This is known as a hub ring. It
helps prevent the disk from being bent or warped in this area of maximum stress.
During the manufacturing process the disk is coated on both sides with a magnetic
material similar to that used on cassette tapes. The coating is burnished in the
read/write area to create a smooth surface. After burnishing, the diskette is tested to
ensure that there are no serious flaws in the coating.
13.2 Which Diskettes to Buy
Floppy diskettes are manufactured in several different sizes and grades. Here is
what you need to tell your dealer when you want to purchase some for use with your
1541 disk drive:
Size: 5 % inch The inner disk is 5 % inches in diameter.
Sectoring: Soft sectored There is one hole in the disk which can be
seen through the index hole when the disk
is rotated once.
Density: Single density Between 4100 and 5800 bits of data are
recorded per inch along a track. This is
slightly more than the normal rating for
a single density diskette but usually
single density diskettes work adequately.
292
Sidedness: Single-sided
Data is recorded only on one side of the
diskette (the back side!).
You can use virtually any 5 % inch diskette in your 1541. The sectoring, density and
sidedness of the diskette are not critical. Note that using a premium quality diskette
will not increase the speed or storage capacity of your drive. As a result, use double-
sided/double density diskettes only if you can get a bargain. Do not pay a premium
for these diskettes!
Most major brands of diskettes will provide adequate service. In general, you get
what you pay for. The magnetic coating on cheap off-brands may be too soft and gum
up your drive's read/write head, or too hard and wear the head away. You are not
really interested in having your valuable programs or data lost!
13.3 The Back Side Controversy
If you cut a new write protect notch on the other edge of the diskette and insert the
diskette into the drive upside down, you can use the second side of the diskette for
storing programs or data. A floppy diskette that has been modified this way is often
called a flippy.
new
write protect
notch
original
write protect
notch
mouse bites
Some companies even sell (for an exorbitant price) a special punch for converting
floppies into flippies. You don't need any special equipment to do this job. Use another
diskette turned upside down and a pencil to mark where the notch should go. Then
use a one-hole paper punch or a pair of scissors to cut out the new notch.
Although you may save some money by using flippies, you may be in for some
long-term problems. Problems can occur for at least three different reasons:
293
1. Flawed surfaces: Most manufacturers of single-sided diskettes guarantee that the
designated recording surface (the back side) is free of any manufacturing defects.
The other side is untested. It may be perfectly OK or it may contain flaws. Before you
store any data on the reverse side of a diskette you should format it and then check it
for flaws using the CHECK DISK program on your 1541TEST/DEMO diskette.
2. Premature wear: Sometimes a flippy will wear out rapidly. One source of wear is
the pressure pad that presses against the non-recording surface of the disk to hold the
disk tightly against the record/play head. The pressure pad is made of a felt-like
material. If it gets contaminated with dirt or grit it can abrade the non-recording
surface. Normally this causes no problems. However, when you start recording on
both sides, it can wear away your data.
Another source of wear is the inner liner. It normally traps dust, smoke, and grit and
removes them from the disk surface. However, when you insert a diskette upside
down, the disk now turns in the opposite direction. The dust, smoke and grit tend to
be redeposited onto the disk where they can cause premature wear. Note that even
double-sided diskettes are designed to rotate in only one direction.
3. Print through: When data is being recorded onto the disk surface, the record/play
head is magnetized by an electric current. On a disk drive that is designed to record
data on only one side of the diskette, the magnetic field may be strong enough to affect
data that is recorded on the other magnetic surface.
Conclusion
There is no agreement in the microcomputer community about whether using flippies
is a good idea. Some users report no problems while others report serious difficulties.
Whether or not you can use flippies probably depends on several things: the type of
disk drive you have and how it is adjusted, the brand of floppy diskettes you use, and
the cleanliness of the environment in which you use and store your diskettes. If you
are interested in using flippies, do a lot of experimenting and testing before you store
valuable programs or data on them, and, as always, keep a backup copy of everything
important.
13.4 The Recording Process
Data is read from or written to the diskette surface by the drive's read/write head. In
its simplest form, a read/write head is a C-shaped piece of steel with a coil of copper
wire wrapped around it, as illustrated below. The break in the steel ring is called the
recording gap. The section of the ring around the recording gap is the part that comes
in contact with the diskette surface.
Record/Play
Head ^^
Ring of
Magnetic
Material
294
Write Mode
ZZZZZZZZZZZZZZZZZZ2ZZZZ2ZZZZZZZZZZZZZZZZZZZ2ZZZ2
Writing
Data
When an electric current passes through the coil, the read/write head becomes an
electromagnet. The gap in the ring causes a strong localized magnetic field. Since
this area of the ring is in contact with the diskette, the diskette becomes magnetized.
As the diskette rotates past the head a strip of the surface becomes magnetized. To
record data on the diskette we change the direction of the current through the coil.
This changes the direction of the magnetization of the magnetic strip. If we let the
diskette make one complete rotation without moving the head, we would find that the
magnetized strip has become a circle. We would have recorded one track of infor-
mation on the diskette.
Since computers work in binary (any character is represented by a string of l's and
O's) and there are two possible directions of magnetization (N-S and S-N), you might
expect that data is recorded on the diskette in a relatively straight forward manner,
one direction of magnetization representing a zero bit and the other direction of
magnetization representing a one bit. Unfortunately, life is not that simple. Suppose
we wanted to record 256 null bytes (00000...) in a row. This would produce a very
long strip with no change in the direction of magnetization. Since the speed of the
drive is not absolutely constant, we might have trouble telling whether there were
supposed to be 254, 256 or even 258 null bytes there.
Commodore gets around the timing problem by encoding the data before it is recorded
on the diskette surface. Each 8-bit byte of data is broken into two 4-bit nybbles. Each
4-bit nybble is converted into a 5-bit code. The codes are designed so that no
combination of encoded characters can ever contain more than two consecutive zero
bits. This is called GCR (Group Code Recording). It solves the timing problem.
295
Hex
Nybble
GCR Code
Hex
Nybble
GCR Code
$0
0000
01010
$8
1000
01001
$1
0001
01011
$9
1001
11001
$2
0010
10010
$A
1010
11010
$3
0011
10011
$B
1011
11011
$4
0100
OHIO
$C
1100
01101
$5
0101
01111
$D
1101
11101
$6
0110
10110
$E
1110
11110
$7
0111
10100
$F
1111
10101
Bits of encoded data are sent to the write circuitry at about 250,000 bits per second
(the actual rate depends on which track is being recorded). The write circuitry makes
one final modification to the data before it is recorded. The direction of current
through the record/play head only changes when a one bit is to be recorded. A zero bit
is represented by a lack of change in the direction of magnetization. Although this
sounds pretty complicated, the example below should help clear things up.
Original Byte ($91): 10010001
Two 4-bit nybbles: 1001 0001
After GCR encoding: 11001 01011
As recorded:
i
1
0
0
1
0
1
0
1
L
N
N
S
S
N
N
s
s
N
N
S
S
1 I I ! I I I I I
Clock Pulses
When the data is read back later, every change in the direction of magnetization is
interpreted as a one bit. If there is no change at the next tick of the clock, a zero bit is
assumed. To keep things synchronized the clock is reset each time a one bit is detected.
Actually, you never have to know about how the data is encoded and recorded on disk.
The DOS takes care of all the picky details. However, you are now in a better position
to understand what happens when you turn your drive OFF or ON with a diskette in
it. The magnetic spike can introduce an extra one bit into the bit stream and throw off
the entire decoding process!
13.5 The Sync A/lark
Data is recorded on a track as a continuous bit stream. There are no blank un-
magnetized areas to indicate the beginning or end of a sector. To enable the drive to
locate where it is on a track, special characters called sync marks are used. A sync
mark is simply eleven or more one bits in a row. Most typical sync marks consist of at
least 30 consecutive one bits. The magnetic pattern recorded for a sync mark looks
like this:
NN S S NN 88 NN 88 NN 8
N N S 8 N N 8
Clock Pulses
296
The sync mark is unique because there is no combination of five-bit codes that can
produce more than eight consecutive one bits. The internal electronics of the 1541 are
designed so that whenever a sync mark is read, the 1541's microprocessor is alerted.
Each sector has two sync marks. One at the start of the header block and another at
the start of the data block.
13.6 How a Sector is Organized
Each sector consists of two parts, a header block and a data block. The header block
contains identifying information. The data block holds the 256 bytes of data that are
stored in this sector. The general layout of a sector is indicated below:
SYNC
MARK
Header Block
Information
Header
Gap
SYNC
MARK
Data Block
Information
Tail
Bap
SYNC
MARK
Header of
next sector
The header block begins with a sync mark. The sync mark warns the drive that there
is important information coming. Next comes the actual header block information. It
includes the track number, sector number and the two-character disk ID. This allows
the drive to identify which of the 683 sectors on the diskette is being read. The final
part of the header block is the header gap. This consists of nine non-sync characters.
These characters are never read and interpreted by the drive. They are there to allow
the drive a bit of time to interpret the header information and get ready to read the
data block.
The data block also begins with a sync mark. It is followed by the data block
information. This includes the 256 data bytes that are stored in the sector. The final
part of the data block is the tail gap. The length of this gap varies but tends to be from
eight to 20 bytes long. Like the characters in the header gap these characters are
never read and interpreted by the drive. They simply allow a bit of breathing space
between sectors.
When you issue a full NEW command, the disk operating system creates and records
a new header block and a dummy data block for each of the 683 sectors on the
diskette. This erases any information that was previously stored on the diskette. Once
these header blocks and dummy data blocks have been recorded, the diskette is said
to be formatted. As you store new files on the diskette, the dummy data blocks
(including the starting sync marks) are overwritten. The header blocks, however, are
never rewritten. They are permanent markers on the diskette until such time as you
reformat the diskette.
Now that you have a general idea of how the parts of a sector are arranged and used,
let's examine the parts in a bit more detail. We'll start with the header block. The
characters that make up a header block are arranged as follows (not as indicated on
page 54 of your 1541 User's Manual):
SYNC
MARK
Header Block
Identifier
Header
Checksum
Sector
Number
Track
Number
Disk ID
Byte #2
Disk ID
Byte #1
Header
297
Header block identifier: The sync mark that begins a header block is always
followed by a $08 character. This indicates that this is a header block and not a data
block. If this character is not $08, the drive will report a 20, READ ERROR, when
you try to access this sector.
Header checksum: This character is used for error checking. The ASCII value of
this character may be found by EORing the sector number, the track number and the
ASCII values of the two disk ID characters. To EOR two numbers convert them to
their eight-bit binary form. Now combine them bit by bit according to the following
rules: 0+0=0, 1+0=1, 0+1=1, and 1+1=0. This is not the same as adding, as there is no
carry; 1+1=0, not 1+1=10. The result may then be combined with the next number.
Let's look at an example:
Find the header checksum for:
Track = 18 Track = 18 = $12 = 00010010
Sector = 1 Sector = 1 = $01 = 00000001
ID = AB (65X66)
EORofl8&l= 00010011
ID #1 (A) = 65 = $41= 01000001
EOR of 18, 1 & 65 = 01010010
ID #2 (B) = 66 = $42 = 01000010
EOR of 18, 1, 65 & 66 = 00010000 = $10
When a header block is read, a new checksum is computed and compared with the
checksum read from the diskette. If they don't match, you get a 27, READ ERROR.
Sector number: Sectors are numbered sequentially from zero to the maximum
number for that track. The maximum sector number for the various tracks are:
Track Range Sector Numbers
1 to 17 0 to 20
18to24 0tol8
25to30 0tol7
31to35 0tol6
Track number: Tracks are numbered sequentially from one to 35. Track 1 is the
outermost track (farthest from the center). Track 35 is the innermost.
Disk ID: This is the two-character disk ID that you specified when the diskette was
formatted. However, their order is reversed. If you specify an ID of AB in a full NEW
command, the headers will contain BA. Since these two characters are recorded as
part of each sector header on the diskette, they are often called the embedded disk ID.
The disk ID characters that you see when you list a diskette's directory are only a
cosmetic ID that is stored on track 18, sector 0. DOS protected diskettes may have
different cosmetic and embedded ID's. On some commercial diskettes different tracks
have different embedded ID's.
298
Now that you have an idea of how a header block is arranged, lef s examine a data
block. It is arranged as follows:
SYNC
MARK
Data Block
Identifier
256 Bytes of
Data
Data
Checksum
Tail
Gap
Data block identifier: The sync mark that begins a data block is always followed by
a $07 character. This indicates that this is a data block and not a header block. If this
character is not $07, the drive will report a 22, READ ERROR, when you try to
access this sector.
Data bytes: Each sector stores 256 bytes of data. In a normal file, the first two bytes
are used to store the track and sector numbers of the next block in the file (a forward
pointer; see Section 13.9). As a result, only 254 bytes of actual data are stored in each
sector of a file.
Data checksum: This character is used for error checking. The ASCII value of this
character may be found by EORing the ASCII values of all 256 data bytes together.
When the data block is read, a new checksum is computed and compared with the
checksum read from the diskette. If they don't match, you get a 23, READ ERROR.
Now that you have an idea of how the various parts that make up a sector are
arranged, let's take a look at a diskette's directory.
13.7 How the Directory Is Organized
The directory is the most important part of any diskette. It is stored on track 18. It
consists of the diskette's Block Availability Map (BAM) on track 18, sector 0 and a file
of directory entries, one for each file stored on the diskette.
In this section we will be examining the directory entries of the 1541TEST/DEMO
diskette. It will be instructive for you to use your computer to follow along. So get out
your copy of the 1541TEST/DEMO diskette. You can use either the DISPLAY T&S
program (from the 1541TEST/DEMO diskette) or the EDIT T&S program (from
Appendix E) to view the various sectors we will be discussing. Turn on your system
and let's get going.
NOTE:
If you have a Commodore 64, you may want to modify the DISPLAY T&S program
on your 1541TEST/DEMO diskette to take advantage of your 40 column screen. To
modify the program, list lines 438 to 448 and make the changes indicated below. Do
not change or delete lines 442 and 446.
438 FOR J=0 TO 31: IF J=16 THEN GOSUB 710
:IF Z*="N"THEN J=80:G0T0 458
440 FOR I=K TO 7
444 IF K=l AND I<6 THEN NB<2)=ASC<A*(I) )
448 A*= " " : B*= " : " : N= J *8 : GOSUB 790 : A*= A*+ "
299
Let's begin by looking at track 18, sector 0. This sector contains the diskette's BAM,
name and cosmetic ID. Here is what you will see when you display this sector. Note
that when you display this sector on your video screen you can only see half the sector
at a time.
oo
08
10
18
20
28
30
38
40
48
50
58
60
68
70
78
80
88
90
98
AO
A8
BO
B8
CO
C8
DO
D8
EO
E8
FO
FB
TRACK 18 SECTOR O
12 Ol 41 00 15 FF FF IF ..A Link, DOS type, BAM track 1
15 FF FF IF 15 FF FF IF BAM tracks 2 and 3
15 FF FF IF 15 FF FF IF BAM tracks 4 and 5
15 FF FF IF 15 FF FF IF BAM tracks 6 and 7
15 FF FF IF 15 FF FF IF BAM tracks 8 and 9
15 FF FF IF 15 FF FF IF BAM tracks 10 and 11
15 FF FF IF 15 FF FF IF BAM tracks 12 and 13
11 D7 5F IF 00 00 00 00 BAM tracks 14 and 15
00 00 00 00 OO OO 00 OO BAM tracks 16 and 17
10 EC FF 07 OO OO 00 00 BAM tracks 18 and 19
00 00 00 00 12 BF FF 07 BAM tracks 20 and 21
13 FF FF 07 13 FF FF 07 BAM tracks 22 and 23
13 FF FF 07 12 FF FF 03 BAM tracks 24 and 25
12 FF FF 03 12 FF FF 03 BAM tracks 26 and 27
12 FF FF 03 12 FF FF 03 BAM tracks 28 and 29
12 FF FF 03 11 FF FF 01 BAM tracks 30 and 31
11 FF FF 01 11 FF FF 01 BAM tracks 32 and 33
11 FF FF 01 11 FF FF 01 BAM tracks 34 and 35
31 35 34 31 54 45 53 54 1541TEST Disk name
2F 44 45 4D 4F AO AO AO /DEMO...
AO AO SA 58 AO 32 41 AO ..ZX.2A. Cosmetic disk ID and
AO AO AO 00 00 00 00 00 DOS type
00 OO 00 00 00 OO 00 00
00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00
00 00 OO 00 00 00 00 00 Garbage <»AB-*FF>
00 OO 00 00 00 00 00 00
00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00
00 OO 00 00 00 00 00 00
00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00
Although the annotations down the side of the display point out the major features,
let's examine the display carefully:
Byte# OO 01 02 03 04 05 06 07
. 00: 12 01 41 00 15 FF FF IF ..A Link to next directory sector
Bytes zero and one are the track ($12=18) and sector ($01=1) numbers of the next
sector in the directory file. These two bytes are ignored when you load a directory, but
are used to locate the next sector when you open the directory file to read it.
Byte# 00 01 02 03 04 05 06 07
. 00: 12 01 41 00 15 FF FF IF ..A DOS type
Byte two is the DOS type ($41=65="A"). If this byte is changed, you will no longer be
able to write to the diskette. You will always get a 73, DOS MISMATCH, error. If you
300
have a dual disk drive you will find that you can no longer duplicate or backup the
diskette either. Some early, simple protection schemes simply modified this byte to
prevent you from copying the diskette.
Byte# 00 01 02 03 04 05 06 07
. 00: 12 01 41 00. 15 FF FF IF ..A Unused byte
Byte $03 is not used by the DOS. It is normally $00 but may be any value.
Byte* 00 01 02 03 04 05 06 07
. 00: 12 01 41 00 15 FF FF 1FJ ..A BAMfortrackl
Bytes $04 to $07 are the BAM for track 1. Since the BAM entry for each track has the
same form, let's examine these bytes in some detail. The BAM entry for a track
consists of a master count of the blocks free for that track and a bit map showing the
status of the individual sectors.
Byte# 00 01 02 03 04 05 06 07
. 00: 12 01 41 00 15: FF FF IF ..A Blocks free on track 1
Byte $04 gives the number of blocks free on track 1. In this case we have $15 or 21
blocks free. None of the blocks on track 1 are in use.
Byte# 00 01 02 03 04 05 06 07
. 00: 12 Ol 41 00 15 FF FF IF ..A..
Bit map showing sector status
Bytes $05 to $07 indicate which sectors are in use. Each bit represents the status of
one sector. If the bit that corresponds to a particular sector is one, that sector is free. If
the bit is zero, the sector is in use. The diagram below indicates which bit represents
which sector:
Byte#
*05
*06
*07
Bit position
76543210
76543210
76543210
Sector number
76543210
111111
54321098
21111
09876
Value for track 1 *FF
Binary value 11111111
Sector status pppppppp
(F=Free, A=Allocated)
*FF
11111111
FFFFFFFF
*1F
00011111
FFFFF
(no sectors 21-23)
301
An example should help make things clear. What is the current status of sector 12 of
track 1? From the diagram we see that sector 12 is represented by bit four of byte
$06. The value stored in byte $06 is $FF (11111111). Bit four is a one; therefore track
1, sector 12 is free.
To be sure that you understand how things work, let's work through another example
for track 14. What is the present status of sector 15 on track 14? From the annotations
down the side of our listing we see that bytes $38 to $3B represent the BAM entry for
track 14.
Byte# 38 39 3A 3B 3C 3D 3E 3F
. 38: iHf D7 5F IF 00 00 00 00 .. Blocks free on track 14
Byte $38 gives the number of blocks free on track 14. In this case we have $11 or 17
blocks free. Some of the blocks/sectors on track 14 are in use. Let's check sector 15.
Byte#
*39
*3A
*3B
Bit position
76543210
76543210
76543210
Sector number
76543210
111111
54321098
21111
09876
Value for track 14
Binary value
Sector status
(F=Free, A=Allocated)
*D7
11010111
FFAFAFFF
*5F
01011111
AiAFFFFF
*1F
0001 1111
FFFFF
Since the bit that corresponds to sector 15 on track 14 is a zero, this sector is in use.
The next part of track 18, sector 0 contains the diskette name.
. 90: 31 35 34 31 54 45 53 54 1541TEST
. 98: 2F 44 45 4D 4F AO AO AO /DEMO... Diskette name
This is the name specified in the NEW command when the diskette was formatted.
Note the three $A0 characters that follow 1541TEST/DEMO. These are shifted
blanks. They are used to pad out the diskette name to 16 characters. You can use
either the EDIT T&S program (Appendix E) or the MOD DISK NAME program
(Appendix E) to change the name stored here.
The cosmetic diskette ID is stored at location $A2 and $A3.
Byte# AO Al A2 A3 A4 A5 A6 A7
. AO: AO AO 5 A 3 AO 32 41 AO . -81. 2 A. Cosmetic diskette ID
302
This is the two-character diskette ID you specified when the diskette was formatted.
This is known as the cosmetic ID because it is only used to generate a directory
listing. The DOS uses the ID embedded in the header blocks to identify a diskette.
The cosmetic DOS type (2A) is stored at location $A5 and $A6.
Byte# AO Al A2 A3 A4 AS A6 A7
. AO: AO AO 5A 58 AO 32 41 AO . . ZX.fi - Cosmetic DOS type
This is known as the cosmetic DOS type because it is only used to generate a directory
listing. The DOS uses the byte $02 of track 18, sector 0 to identify the diskette's DOS
type.
The remainder of track 18, sector 0 is essentially garbage. Some disk protection
schemes have even used this area for storing machine language programs!
This completes our tour of track 18, sector 0. Now let's take a look at the actual
directory entries themselves. They begin on track 18, sector 1.
Track 18, sector 1 is the start of the file of directory entries. Since the file entries are
the key to finding our way around the diskette, let's take a look at them in some detail.
00
08
10
18
20
28
30
38
40
48
50
58
60
68
70
78
80
88
90
98
AO
A8
BO
B8
CO
TRACK 18 SECTOR 1
12 04 82 11 00 48 4F 57 HOW Entry 1: file type = $82 (130)
20 54 4F 20 55 53 45 AO .TO. USE. First block on $11/$00 (17/0)
AO AO AO AO AO 00 00 00 File name = HOW TO USE
00 00 00 00 00 00 OD 00 Length = $0D (13) blocks
00 00 82 11 03 48 4F 57 HOW Entry 2: file type = $82 (130)
20 50 41 52 54 20 54 57 .PART.TW First block on $ll/$03 (17/3)
4F AO AO AO AO 00 00 OO O File name = HOW PART TWO
00 00 00 00 00 00 05 00 Length = $05 (5) blocks
00 00 82 11 09 56 49 43 VIC Entry 3: file type = $82 (130)
2D 32 30 20 57 45 44 47 -20.WEDG First block on $ll/$09 (17/9)
45 AO AO AO AO 00 00 00 E File name = VIC-20 WEDGE
00 00 00 00 00 00 04 00 Length = $04 (4) blocks
00 00 82 13 00 43 2D 36 C-6 Entry 4: file type = $82 (130)
34 20 57 45 44 47 45 AO 4. WEDGE. First block on $13/$00 (19/0)
AO AO AO AO AO 00 00 00 File name = C-64 WEDGE
00 00 00 00 OO 00 01 00 Length = $01(1) block
00 00 82 13 01 44 4F 53 DOS Entry 5: file type = $82 (130)
20 35 2E 31 AO AO AO AO .5.1 First block on $13/$01 (19/1)
AO AO AO AO AO 00 00 00 File name = DOS 5.1
00 00 00 OO 00 00 04 00 Length = $04 (4) blocks
00 OO 82 13 03 43 4F SO COP Entry 6: file type = $82 (130)
59 2F 41 4C 4C AO AO AO Y/ALL. . . First block on $13/$03 (19/3)
AO AO AO AO AO 00 00 00 File name = COPY/ALL
00 00 OO 00 00 00 OB 00 Length = $0B (11) blocks
00 00 82 13 09 50 52 49 PRI Entry 7: file type = $82 (130)
303
C8: 4E 54 43 52 20 54 45 53 NTER.TES First block on $13/$09 (19/9)
DO: 54 AO AO AO AO 00 00 00 T File name = PRINTER TEST
DB: 00 00 00 00 00 00 09 00 Length = $09 (9) blocks
EO: 00 00 82 10 00 44 49 53 DIS Entry 8: file type = $82 (130)
E8: 4B 20 41 44 44 52 20 43 K.ADDR.C First block on $10/$00 (16/0)
FO: 48 41 4E 47 45 00 00 00 HAN6E. . . File name = DISK ADDR CHANGE
F8: 00 00 00 00 00 00 04 00 Length = $04 (4) blocks
The annotations down the side of the listing point out the major features of the eight
entries. Let's examine the first part of the listing in some detail.
Byte* 00 01 02 03 04 05 06 07
. 00: 12 04 82 11 00 48 4F 57 HOW Link to next directory sector
Bytes $00 and $01 of this sector ($12=18; $04=4) are not really part of any entry. They
are the track and sector numbers of the next sector in the directory (18/4). Note that
the other seven entries in this sector have $00/$00 in this position.
Byte $02 is really the start of the first directory entry.
Byte# 00 01 02 03 04 05 06 07
. 00: 12 04 82 11 00 48 4F 57 HOW Entry 1: file type = $82 (130)
This is the file type byte. The value of this byte determines the file type. The table
below summarizes the major file types:
Hex
Decimal
File Type
Comment
$00
0
Scratched
Not shown in a directory listing.
$01
1
*SEQ
An unclosed sequential file.
$02
2
*PRG
An unclosed program file.
$03
3
*USR
An unclosed user file.
$80
128
DEL
An unusual, but legal, file type.
$81
129
SEQ
A normal sequential file.
$82
130
PRG
A normal program file.
$83
131
USR
A normal user file.
$84
132
REL
A normal relative file.
$C0
192
DEL<
A locked unusual file type.
$C1
193
SEQ<
A locked sequential file.
$C2
194
PRG<
A locked program file.
$C3
195
USR<
A locked user file.
$C4
196
REL<
A locked relative file.
NOTES:
1. Locked files cannot be scratched. The file type byte must be changed back to a
normal value before the file can be scratched. See Section 13.8.
304
2. By editing the file type byte, unusual file types can be produced (see Section 13.8).
Byte# OO 01 02 03 04 05 06 07
. 00: 12 04 82 1 1 ' O 48 4F 57 HOW First block on $ll/$00 (17/0)
Bytes $03 and $04 give the track and sector numbers of the sector where the first
block of the file is stored. In this case the first block of the file is on track 17, sector 0.
Byte# 00 01 02 03 04 05 06 07
. 00: 12 04 82 11 00 48 4F 57 . .... HOW File name = HOW TO USE
Byte* 08 09 OA OB OC OD OE OF
. 08: 20 54 4F 20 55 53 45 AO .TO. USE.
Byte# 10 11 12 13 14 15 16 17
Bytes $05 to $14 contain the file name. Note that the name is padded with shifted
blanks ($A0=160) so that the name is always 16 characters long.
Byte# 10 11 12 13 14 15 16 17
. 10: AO AO AO AO AO Of >0 > REL files only
Bytes $15 to $17 are only used with relative file entries. They are always $00 for other
file types. Bytes $15 and $16 hold the track and sector numbers of the side-sector file
for the relative file. Byte $17 holds the record size. More details about these bytes may
be found in Section 13.13 on relative file storage.
Bytett 18 19 1A IB 1C ID IE IF
. 18: 00 00 00 00 00 00 OD 00 Unused bytes
Bytes $18 to $1B are unused. They are always $00.
Byte# 18 19 1A IB 1C ID IE IF
. 18: 00 00 00 00 00 '.>0 OD 00 Used during replacement only
Bytes $1C and $1D are used only temporarily during a replace operation (SAVE
replacement or OPEN to replace). During these operations they hold the track and
sector numbers of the first block in the new replacement file. Otherwise they are
always $00.
305
Byte# 18 19 1A IB 1C ID IE IF
. is: oo oo oo oo oo oo
Length = $0D (13) blocks
The final two bytes in a directory entry $1E and $1F give the length of the file in
blocks. The file length is in lo byte/hi byte form. To find the length of the file, multiply
the value of byte $1F by 256 and add the value of byte $1E like this:
Length of file = 256 x ($1F value) + ($1E value)
= 256x(0) + (13)
= 13 blocks
All the other file entries in the directory have this same form. To acquire facility in
reading the directory, you should spend some time identifying the parts of the other
entries.
To locate the next sector of directory entries simply follow the pointer in bytes $00
and $01. In this case the next sector will be $12/$04, track 18 sector 4. Let's take a
brief look at this sector.
TRACK 18 SECTOR 4
.
OO
: oo
FF
82
10
01
44
.
08
: AO
AO
AO
AO
AO
AO
■
10
: ao
AO
AO
AO
AO
00
■
18
: oo
OO
00
00
00
00
.
20
: oo
OO
82
10
03
56
.
28
: 37
20
42
41
40
AO
■
30:
AO
AO
AO
AO
AO
00
■
38
: oo
00
00
00
00
00
■
40:
00
00
82
10
07
43
■
48:
: 43
4B
20
44
49
53
■
50:
: AO
AO
AO
AO
AO
00
.
38.
: oo
00
00
00
00
00
m
60:
00
OO
82
10
OF
44
.
68
. so
4C
41
59
20
54
■
70:
AO
AO
AO
AO
AO
OO
■
78"
: oo
00
00
00
00
00
■
80:
00
00
82
14
02
50
■
88:
: 46
4F
52
4D
41
4E
■
90:
20
54
45
53
54
00
.
98
: oo
00
00
00
00
00
■
ao:
: oo
00
82
14
07
53
■
A8
: 55
45
4E
54
49
41
■
BO
: 46
49
4C
45
AO
00
.
B8
: oo
00
00
00
00
00
.
CO
: oo
00
82
OF
01
52
■
C8
: 44
4F
4D
20
46
49
■
DO
: AO
AO
AO
AO
AO
00
.
D8
: oo
00
00
00
00
00
■
EO
: oo
OO
00
00
00
00
■
E8
: oo
00
00
00
00
00
■
FO
: oo
00
00
00
OO
00
.
F8
: oo
00
00
OO
OO
00
49 52 D1R Entry 1: file type = $82 (130)
AO AO First block on $10/$01 (16/1)
00 00 Pile name = DIR
04 00 Length = $04 (4) blocks
49 45 VIE Entry 2: file type = $82 (130)
AO AO W.BAM... First block on $10/$03 (16/3)
00 00 File name = VIEW BAM
06 00 Length = $06 (6) blocks
48 45 CHE Entry 3: file type = $82 (130)
4B AO CK.DISK. First block on $10/$07 (16/7)
00 00 File name = CHECK DISK
04 00 Length = $04 (4) blocks
49 53 DIS Entry 4: file type = $82 (130)
26 53 PLAY.T&S First block on $10/$0F (16/15)
00 OO File name = DISPLAY T&S
OE 00 Length = $0E (14) blocks
45 52 PER Entry 5: file type = $82 (130)
43 45 FORMANCE First block on $14/$02 (20/2)
00 00 .TEST... File name = PERFORMANCE TEST
09 00 Length = $09 (9) blocks
45 51 SEQ Entry 6: file type = $82 (130)
4C 20 UENTIAL. First block on $14/$07 (20/7)
00 00 FILE.... File name = SEQUENTIAL FILE
05 00 Length = $05 (5) blocks
41 4E RAN Entry 7: file type = $82 (130)
4C 45 DOM. FILE First block on $0F/$01 (15/1)
00 00 File name = RANDOM FILE
OD 00 Length = $0D (13) blocks
00 00
00 00
00 00
00 00
306
Although this display is similar to the one for sector 1, there are two things to notice.
The most obvious difference is that this sector of the directory is not full. It contains
only seven file entries. The last entry is blank. When a new file is stored on this
diskette its directory entry will be stored from $E0 to $FF. The second thing to notice
is the pointer to the next directory sector (bytes $00 and $01). The pointer says that
the next sector of the directory is on track 0 ($00), sector 255 ($FF). This seems
strange. There is no track 0, sector 255! Actually, it means that this is the last sector
in the directory. Whenever you see a pointer to track 0 it means that you have reached
the end of the file you have been tracing.
This concludes our tour of the directory. Now that we know what the various parts of
the directory do, let's create some non-standard directory entries.
13.8 Modifying the Directory
Now that you understand how the directory is organized, you are in a position to be
able to modify the directory. By editing track 18, sector 0 you can change a diskette's
name. By editing a file's directory entry you can lock or unlock the file, create a
strange file name or an unusual file type. You can even change the length of the file so
that it appears to be thousands of blocks long.
There are three programs listed in this book that you can use to modify a diskette's
directory: EDIT T&S (Appendix E), MOD DISK NAME (Appendix E), and MOD
ENTRY (listed below). The EDIT T&S program is the most powerful program. It
allows you to edit any part of the directory. However, you must understand the
structure of the directory thoroughly to use this program! Don't practice on your only
copy of a diskette! The MOD DISK NAME is a short, special purpose utility that
allows you to change a diskette's name quickly and easily. The MOD ENTRY program
makes it very easy to change a directory entry's file type, file name and length. It also
shows you what effect the change has had on a directory listing. The MOD DISK
NAME and MOD ENTRY programs do not allow you to create names that contain
special characters (nulls, clear screen characters, etc.). You will have to use the EDIT
T&S program to do this.
The MOD DISK NAME program is easy to use. Insert the diskette whose name is to
be changed and RUN the program. The existing diskette name will be displayed;
Make any changes you like. The new name will then be recorded on the diskette.
When you load and run the MOD ENTRY program, you will be asked if you want to
work with file names or directory positions. If you choose the name option, you will be
asked for the name of the file entry you want to modify. Just type in the name (wild
cards are OK) and press RETURN. If you choose the position option you will be asked
for the number of the directory sector that contains the entry (1, 4, 7,...) and the
entry's position in the sector (0-7). This option is provided so that you can rescue a file
which no longer appears in a directory listing. If you are using a 1541 drive, you can
use either option. If you are using a 4040 drive you can only use the position option.
Once you have identified which file you want, the directory entry for that file will be
displayed. The display of the entry for the HOW TO USE file on your 1541TEST/
DEMO diskette looks like this:
307
MODIFY DIRECTORY ENTRY
FILE NAME - HOW TO USE
FILE TYPE BYTE ■ 130
TYPE NORMAL LOCKED UNCLOSED
DEL
128
192
0
SEQ
129
193
1
PR6
130
194
2
USR
131
195
3
REL
132
196
4
FILE LENGTH - 13
DIRECTORY ENTRY
0 "1541 TEST /DEMO " ZX 2A
13 "HOW TO USE" PRG
558 BLOCKS FREE
EDIT THIS ENTRY <Y/N>?
If you type N and press RETURN, the program will terminate (to edit a different
entry, simply rerun the program). However, if you respond with a Y, editing begins.
A question mark will appear on the screen right next to the name field like this:
FILE NAME =? HOW TO USE
When you have finished making changes to the file's name, press RETURN. You will
then have a chance to edit the file type and length. If you don't want to make any
changes to a particular field, simply press RETURN.
Once all three fields have been entered, you will be asked whether you want to re-edit
the entry.
RE-EDIT THIS ENTRY <Y/N>? _
If you respond with a Y, you have another chance to make changes. Once you are
satisfied with the entry, respond with an N.
You will then be asked if you want to have this entry recorded in the directory.
CHANGE ON DISKETTE <Y/N>? _
If you type N and press RETURN, the program will terminate. However, if you
respond with a Y, the modified entry will be recorded in the diskette's directory. In a
few moments, the entry will be redisplayed and you will get to see what the entry will
look like in the directory listing.
308
CAUTION:
Modifying directory entries can be fun and interesting. However, you may manage to
make some or all of the files inaccessible. Use a test diskette for practice and
experimentation.
Now that you know how to use the MOD ENTRY program, try some experiments.
What happens when you try to SCRATCH a locked file? What file type bytes give you
the file types SR?, EQ? and G? Can you load or open any of these strange file types?
Here are a few more things you can try:
1. What happens to the directory when you include a null character in the file name
like this?
SAVE CHR*<0)+"-filename",8
2. What happens if you put a shifted blank in the file name?
SAVE " GOOD " +CHR* ( 1 60 ) + " STUFF " , 8
3. Here's another one to try. But be sure to remember the hidden letter if you ever
want to load it again.
SAVE "H"+CHR* <20) +"-f i lename" , 8
4. Here's a good way to disguise a program. Change it from a PRG file into a DEL
file by editing the file type byte. Now try this:
LOAD "-filename, DEL, R", 8
5. Have you ever thought of time-stamping the various versions of the program you
are working on? Try this:
SAVE "filename "+TI*,8
The possibilities for playing with the directory are endless. Enjoy!
NOTE:
If you are a programmer, you may want to study some of the techniques used in the
MOD ENTRY program. The techniques which are used for editing, accessing a
directory entry and reading a selective directory listing may be of interest.
******************
PROGRAM: MOD ENTRY
******************
0 REM 1541 USER'S GUIDE SECTION 13.8
1 REM COPYRIGHT: G. NEUFELD, 1984
2 :
3 REM MODIFY FILE'S NAME, TYPE OR LENGT
H
4 :
100 REM MOD ENTRY
110 PRINT" {CLRJ MOD I FY DIRECTORY ENTRY"
120 GET A*: IF A*<>"" GOTO 120
130 INPUT" CDOWN>USE NAME OR POSITION (N/
P) N CLEFT 3J";X*
140 IF LEFT*<X*, 1)<>"P" THEN ZF=0:GOTO 2
80
150 :
160 REM WORK FROM SECTOR* AND ENTRY NUMB
ER
170 ZF=1
180 INPUT" CDOWN> DIRECTORY SECTOR NUMBER
1 CLEFT 3>";X*
190 SE=VAL<X*):IF SE<1 OR SEM8 GOTO 180
200 INPUT" CDOWN>POSITION IN SECTOR (0-7)
0O.EFT 33" IX*
210 P=INT(VAL(X*)):IF P<0 OR P>7 GOTO 20
O
220 PT=2 + 32*P
230 OPEN 15,8, 15, "IO"
240 OPEN 2,8,4, "#0"
250 GOTO 500
260 :
270 REM WORK FROM NAME
280 PRINT" CCLR>MODIFY DIRECTORY ENTRY"
290 PRINT" <DOWN>FILE NAME "F*
300 PRINT"<:UP>"TAB(11) ;: INPUT F*
310 :
320 OPEN 15,8, 15," 10"
330 OPEN 1,8,5,F*+",M"
340 INPUT#15,EN,EM*,ET,ES
350 IF EN<20 GOTO 400
360 PRINT" {DOWN* DISK ERROR"
370 printen?em*;et;es
380 CLOSE l: CLOSE 15: END
390 :
400 OPEN 2,8,4, "#0"
410 IF ZF=1 GOTO 500
420 :
430 REM DIG OUT SECTOR & POINTER TO ENTR
Y
440 PRINT#15, "M-R"CHR* < 144) CHR* <2>
450 GET# 1 5 , A* : SE= ASC ( A*+CHR* ( O ) )
460 PRINT#15, "M-R"CHR* ( 148) CHR* (2)
470 GET# 1 5 , A* : PT=ASC ( A*+CHR* ( O ) )
480 :
310
490 REM READ SECTOR AND DIG OUT ENTRY
500 PRINT#15, "Ul:4 O 18"SE
510 PRINT#15, "B-P:4"PT
520 GET#2 , A* : FT=ASC ( A*+CHR* < O ) >
530 GET#2,TL*,SL*
540 F*=""
550 FOR K=l TO 16
560 GET#2 , A* : I FA*= " " THEN A*=CHR* ( O )
570 F*=F*+A*
580 NEXT
590 PR I NT# 15," B-P : 4 " PT+28
600 GET#2 , A* : FL= ASC < A*+CHR* ( O ) )
6 1 O GET#2 , A* : FL=FL+256*ASC ( A*+CHR* ( O ) )
620 CLOSE 1: CLOSE 2
630 :
640 REM DISPLAY ENTRY
650 PRINT" {CLR} MOD I FY DIRECTORY ENTRY"
660 PRINT" CDOWN} FILE NAME = ";F*
670 PRINT" {DOWN} FILE-TYPE BYTE = ";FT
680 PRINT" {DOWNMRVS} TYPE NORMAL LOCKED
UNCLOSED"
690 PRINT" DEL 128 192 O"
700 PRINT" SEQ 129 193 1"
710 PRINT" PRG 130 194 2"
720 PRINT" USR 131 195 3"
730 PRINT" REL 132 196 4"
740 PRINT" {DOWN} FILE LENGTH = "FL
750 PRINT" {DOWN* DIRECTORY ENTRY {DOWN}"
760 :
770 REM DISPLAY WHAT IT LOOKS LIKE IN DI
RECTORY LIST
780 TRY=0
790 OPEN 1,8,0, "*0:"+F*+",P,R"
800 INPUT#15,EN,EM*,ET,ES
810 IF EN<20 GOTO 840
820 TRY=TRY+l:IFTRY<5 THEN CLOSEl:PRINT#
15, " I " : G0T0790
830 PRINT" €DOWN}DISK ERROR":PRINTEN;EM«;
ET ; ES : CLOSE 1 5 : STOP
840 GET#1,A*:6ET#1,A*:F0R K=l TO 3
850 get#1,a*:get#1,a*:ifk=3 then get#1,a
$:get#i,a*
860 get# 1 , a* : l=asc ( a*+chr* ( o ) )
870 get# 1 , a* : l=l+256*asc < a*+chr* < o ) ) : pr i
NTL;
880 FORM= 1 T025 : GET# 1 , A* : PR I NT A* ; : NEXTM
890 GET#1,A»:IF A*<>"" AND K=2 THEN PR IN
TA*;
900 PRINT: NEXT: CLOSE 1
910 :
311
920 REM EDIT ENTRY
930 INPUT" {DOWN} ED IT THIS ENTRY <Y/N) N
{LEFT 3>";X*
940 IF X*<>"Y" THEN CLOSE 15: END
950 PRINT" OJPJ
ll
960 PRINT" <HOME> {DOWN 2>"TAB( 11) ;: INPUT
F*
970 IF LEN(F*)>16 THEN PRINT" <UP> "TAB (29
>" ":GOTO 960
980 PRINT"{D0WN>"TAB(16);: INPUT NT
990 IF NT>=0 AND NT<256 AND NT=INT(NT) T
HEN FT=NT:GOTO 1040
1000 IF NT>255 THEN PRINT " {UP* "TAB (25) "
MAX IS 255"
1010 IF NT<0 THEN PRINT "{UP>"TAB(25) "MI
N IS O"
1020 IF NTOINT<NT) THEN PRINT "{UP* "TAB
<25>" INTEGERS ONLY"
1030 FOR K=l TO 2000: NEXT: GOTO 650
1040 PRINT" CDOWN 8> "TAB ( 13) ;: INPUT NL
1050 IF NL>=0 AND NL< 65536 AND NL=INT(NL
) THEN FL=NL:GOTO HOO
1060 IF NL>65535 THEN PRINT "CUP} "TAB (25
)"MAX IS 65535"
1070 IF NL<0 THEN PRINT " CUPVTAB (25) "MI
N IS 0"
1080 IF NLOINT(NL) THEN PRINT "{UP* "TAB
(25) "INTEGERS ONLY"
1090 FOR K=l TO 2000: NEXT: GOTO 650
1100 PRINT"<DOWN 6> "
lllO INPUT "RE-ED IT THIS ENTRY (Y/N) NCL
EFT 3>";X*
1120 IF LEFT*(X«, 1)<>"N" GOTO 950
1130 INPUT" <UP> CHANGE ON DISKETTE (Y/N)
Y {LEFT 5>";X*
1140 IF LEFT*(X*, 1)<>"Y" THEN CLOSE 15: E
ND
1150 :
1160 REM MODIFY ENTRY ON DISKETTE
1170 IF LEN(F*)<16 THEN F*=F*+CHR* ( 160) :
GOTO 1170
1180 PRINT" OJP>WRITING TO DISKETTE
II
1190 OPEN 2,8,4, "#"
1200 PRINT#15, "Ul:4 O 18"SE
1210 PRINT#15, "B-P:4"PT
1220 PRINT#2,CHR*(FT);
1230 PRINT#15, "B-P:4"PT+3
1240 PRINT#2,F*;
312
1 250 PR I NT# 1 5 , " B-P : 4 " PT+28
1260 H=INT<FL/256>:L=FL-256*H
1270 PRINT#2, CHR* <L> CHR* (H) ;
1280 PRINT#15, "U2:4 O 18"SE
1290 CLOSE 2
1300 ZF=l:REM DON'T LOSE OUR PLACE
1310 GOTO 400
NOTE:
Sections 13.9 through 13.14 deal with how files are stored on a diskette. If you are an
average user of disk files, you do not have to be concerned about the way in which files
are stored. All you have to do is open a file, read or write data, and when you are done,
close it again. You will never see a forward pointer or side sector and you don't have to
worry about how your files are stored. The Disk Operating System (DOS) takes care
of all those details for you.
However, if you are interested or having troubles, read on. You can learn the inner
mysteries of your disk drive.
13.9 How Forward Pointers Work
The key to understanding Commodore file storage is the idea of forward pointers.
Each part of a file contains a pointer that points ahead to the next part. We saw some
forward pointers when we explored the file of directory entries in Section 13.7. The
values stored in bytes $00 and $01 of any block in the directory were the track and
sector numbers of the next block in the file. The table below summarizes the pointers
in the directory file of the 1541TEST/DEMO diskette.
Dir Track/ Value Value Points forward to
Block Sector in $00 in $01 the next block at
#1 18/0 $12(18) $00(1) Track 18, sector 1
#2 18/1 $12(18) $04(4) Track 18, sector 4
#3 18/4 $00(00) $FF(255) Last block in file
All Commodore disk files use forward pointers like this. They make it easy to trace a
file, just find the first block and follow the pointers stored in bytes $00 and $01. When
you find a track link of $00 (zero), you've reached the end of the file. But how do you
find that first block? It's in the directory entry!
Let's do a bit of exploring and see what we can find. We'll trace through the program
HOW PART TWO. Here is the directory entry:
TRACK 18 SECTOR 1
. 20: OO 00 82 11 03 48 4F 57 HOW
. 28: 20 50 41 52 54 20 54 57 .PART.TW
313
30: 4F A0 AO AO AO 00 00 00 O.
38: 00 00 00 00 00 00 05 00
The pointer to the first block in the file is in bytes $23 and $24. The first part of the
program is stored on track $11 (17), sector $03 (3). Let's take a look.
TRACK 17 SECTOR 3
oo:
11
OD
01
04
OA
04
08:
99
22
93
22
00
2D
10:
03
99
22
20
20
20
18:
20
20
20
20
20
20
20:
52
46
4F
52
4D
41
28:
45
20
54
45
53
54
30:
36
04
20
03
99
22
38:
00
62
04
2A
03
99
40:
54
48
45
20
50
45
48:
4F
52
4D
41
4E
43
50:
54
45
53
54
20
50
58:
: 47
52
41
4D
20
41
60:
. 4F
57
53
22
00
8C
68:
: 03
99
22
41
4E
59
70:
: 45
20
54
4F
20
54
78
: 54
20
54
48
45
20
80
: 45
43
54
52
4F
4E
88
: 20
41
4E
44
20
22
90
: 04
3E
03
99
22
4D
98
: 48
41
4E
49
43
41
AO
: 43
41
50
41
42
49
A8
: 54
49
45
53
20
4F
BO
: 54
48
45
20
44
49
B8
: 20
22
00
DA
04
48
CO
: 22
44
52
49
56
45
C8
: 48
45
4E
45
56
45
DO
: 4E
45
43
45
53
53
D8
: 59
2E
20
22
00
08
EO
: 03
99
22
55
53
45
E8
: 48
49
53
20
50
52
FO
: 52
41
4D
20
57
48
F8
: 45
56
45
52
20
59
OC 03
04 16 .".".-..
20 20 .."
50 45 PE
4E 43 RFORMANC
22 00 E.TEST".
20 22 6 "."
22 20 ...*..".
52 46 THE.PERF
45 20 ORMANCE.
52 4F TEST. PRO
4C 4C SRAM. ALL
04 34 0WS"...4
4F 4E .."ANYON
45 53 E.TO.TES
45 4C T. THE. EL
49 43 ECTRONIC
00 B8 .AND."..
45 43 . >. ."MEC
4C 20 HANICAL.
4C 49 CAPABILI
46 20 TIES. OF.
53 4B THE. DISK
03 99 . "...H. .
20 57 "DRIVE. W
52 20 HENEVER.
41 52 NECESSAR
05 52 Y.."...R
20 54 .."USE.T
4F 47 HIS. PROG
45 4E RAM. WHEN
4F 55 EVER. YOU
This may not look like the program listings you are familiar with, but it is a program
just the same. The BASIC words like PRINT, GOTO and FOR are all tokenized. If
you worked through Section 6.7, you may recognize some familiar landmarks such as
line pointers, line numbers, null bytes marking the ends of lines, etc. However, this
time only the first two bytes are of interest.
00:
04 OA 04 OC 03
314
Bytes $00 and $01 are the pointer to the next block in the file. In this case they point to
track 17 ($11), sector 13 ($0D). Let's follow this forward pointer to the next block on
track 17, sector 13. We'll only look at the first few lines of the display because we are
primarily interested in following the file chain pointers.
TRACK 17 SECTOR 13
. 00: 11 05 20 53 55 53 50 45 ...SUSPE
. 08: 43 54 20 22 00 36 05 5C CT. ".6.
. 10: 03 99 22 54 48 41 54 20 . . "THAT.
This time the forward pointer (bytes $00 and $01) indicate that the next block is on
track 17 ($11), sector 5 ($05). Let's take a look.
TRACK 17 SECTOR 5
. 00: 11 OF 20 41 4E 44 20 43 ...AND.C
. 08: 41 4E 22 00 30 06 AC 03 AN".0. ..
. 10: 99 22 42 45 20 55 53 45 ."BE. USE
Nothing of interest here. Let's follow the pointer to track 17 ($11), sector 15 ($0F).
TRACK 17 SECTOR 15
. 00: 11 07 4E 44 4F 4D 20 46 . .NDOM.F
. 08: 49 4C 45 20 42 4F 54 48 ILE.BOTH
. 10: 22 00 33 07 F2 03 99 22 ".3. ..."
Again, nothing new here. Onwards, to track 17 ($11), sector 7 ($07).
00
08
10
18
20
28
30
38
40
48
50
58
60
68
TRACK 17 SECTOR 7
00 52 2E 22 00 04 08 2E .R."
04 8D 39 30 30 30 00 OA ..9000..
08 40 IF 80 00 26 08 28 . ...&.<
23 99 22 20 3C 3C 50 52 #. ".«PR
45 53 53 20 53 50 41 43 ESS.SPAC
45 20 42 41 52 3E 3E 22 E.BAR»"
00 3A 08 32 23 Al 41 24 .:.2#.A*
3A 8B 41 24 B2 22 22 A7 :.A*."".
39 30 31 30 00 46 08 3C 9010. F.<
23 99 22 93 22 3B 3A 8E #.".";:.
00 00 00 45 59 20 41 4C ...EY.AL
53 4F 20 22 00 81 06 CO SO."
03 99 22 49 4C 4C 55 53 .."ILLUS
54 52 41 54 45 20 54 48 TRATE.TH
315
70
78
80
88
90
98
A0
A8
BO
B8
CO
C8
DO
D8
EO
E8
FO
F8
45 20 49 4D 50 4F 52 54 E. IMPORT
41 4E 54 20 54 45 43 48 ANT. TECH
4E 49 51 55 45 20 22 00 NIQUE. ".
AC 06 CA 03 99 22 4F 46 "OF
20 43 48 45 43 4B 49 4E .CHECK IN
47 20 54 48 45 20 45 52 B.THE.ER
52 4F 52 20 43 48 41 4E ROR.CHAN
4E 45 4C 20 41 46 54 45 NEL.AFTE
52 22 00 D3 06 D4 03 99 R"
22 45 41 43 48 20 41 43 "EACH. AC
43 45 53 53 20 54 4F 20 CESS. TO.
54 48 45 20 44 49 53 4B THE. DISK
20 44 52 49 56 45 2E 20 .DRIVE..
22 00 DD 06 D9 03 8D 39 " 9
30 30 30 00 09 07 E8 03 000
99 22 53 45 51 55 45 4E . "SEQUEN
54 49 41 4C 20 46 49 4C TIAL.FIL
45 20 41 4E 44 20 52 41 E.AND.RA
Ah ha! Finally something different. A track link of zero ($00). We've reached the end
of our file! However, this time we don't have a sector link of 255 ($FF) like we had
with the last block in the directory file. This time we have a sector link of 82 ($52).
What does that mean? Let's take a look around byte $52 and see if we can figure it out.
. 40: 39 30 31 30 00 46 08 3C 9010. F.<
. 48: 23 99 22 93 22 3B 3A 8E #.".";:.
. 50: 00 00(^45 59 20 41 4C ...EY.AL
. 58: 53 4F 20 22 00 81 06 CO SO."
. 60: 03 99 22 49 4C 4C 55 53 .."ILLUS
The only thing that looks peculiar is that bytes $50, $51 and $52 all have the value
$00. If you have read through Section 6.7, you should remember that a BASIC
program always ends with three $00 bytes. One to end the last line, and another two
to indicate that there are no more lines. Now we know the meaning of the sector link
in the last block of a file. It points to the last byte in the sector that is part of the file.
NOTE:
Only BASIC program files end with three null bytes. Don't expect to find null bytes
on the end of other types of files.
But what about all the rest of that stuff in the sector from bytes $53 to $FF? It sure
looks like part of a BASIC program. It may well be. However, it is not part of this file.
It's junk. As the program was being saved each block of the file was built up in the
disk drive's RAM before it was encoded and recorded onto the diskette. This time
there wasn't enough program to completely fill the last block so the drive took the end
of the program and whatever else happened to be in that 256 bytes of RAM, encoded
it, and recorded it on the diskette. You'll find many strange things in the last blocks of
files.
316
Here is a program that you can use to trace any file on a diskette. The program is very
easy to use. Just run it and enter the name of the file to be traced (wild cards are OK).
You will see a listing of the entire file chain, including bytes $00 and $01 of the final
block. The program even lists the side-sector file chain for relative files.
*******************
PROGRAM: TRACE FILE
*******************
0 REM 1541 USER'S GUIDE SECTION 13.9
1 REM COPYRIGHT: G. NEUFELD, 1984
2 :
3 REM TRACE & DISPLAY FILE CHAIN FOR NA
MED FILE
4 :
lOO REM TRACE FILE
110 PRINT" £CLR> TRACE FILE 1541"
120 INPUT" CDOWN3 FILE NAME";F*
130 PRINT" {DOWN} FILE CHAIN<DOWN> "
140 OPEN 15,8,15, "IO"
150 OPEN 1,8,4, "#"
160 0PEN2,8,5,F*
170 GOSUB 600
180 :
190 REM DIG OUT POINTER TO DIRECTORY ENT
RY
200 PRINT#15, "M-R"CHR*(144)CHR*<2)
210 GET#15,A*:SE=ASC<A*+Z*>
220 PRINT#15, "M-R"CHR*<148)CHR*<2>
230 GET#15,A*:PT=ASC<A*+Z*>
240 :
250 REM DIG OUT FILE TYPE & POINTER TO M
AIN FILE
260 PRINT#15, "Ul:4 O 18"SE
270 PRINT#15, "B-P:4"PT
280 GET# 1 , A* : FT= ASC ( A«+CHR$ ( O ) )
290 GET# 1 , A* : TL=ASC < A*+CHR* ( O ) )
300 GET# 1 , A* : SL= ASC ( A«+CHR$ ( O ) )
310 REM DIG OUT POINTER TO SIDE SECTOR F
ILE
320 PRINT#15, "B-P:4,,PT+19
330 GET# 1 , A* : RT= ASC ( A*+CHR* < O ) )
340 get# 1 , a* : rs= asc ( a* +chr* ( o ) )
350 print" {down} ma in file": gosub 430: pr i
nt:printm*
360 IF FT0132 GOTO 390
370 tl=rt:sl=rs
380 PRINT" {DOWN* SIDE SECTOR FILE": GOSUB
430 : PR I NT : PR I NTM*
390 PRINT" {DOWN>ALL DONE"
317
400 CLOSE l: CLOSE 2: CLOSE 15: END
410 :
420 REM PRINT LINK AND CHECK IF OK
430 NB=0
440 PRINT RIGHT* <STR*<TL),2>","RIGHT*("0
"+MID*(STR*<SL),2),2>" "5
450 NB=NB+l:IF NB/6 = INT<NB/6> THEN PRI
NT
460 IF TL=0 THEN M*="END OF FILE": RETURN
470 IF TL>35 THEN M*="BAD TRACK LINK": RE
TURN
480 NS = 20+2*(TL>17)+<TL>24)+<TL>30)
490 IF SL>NS THEN M*="BAD SECTOR LINK":R
ETURN
500 :
510 REM READ IN NEXT BLOCK AND GET NEW L
INK
520 PRINT#15,"Ul:4 0"TL?SL
530 GOSUB 60O
540 PR I NT# 15," B-P : 4 O "
550 GOSUB 600
560 GET# 1 , A* : TL= ASC ( A*+CHR* ( O > )
570 GET* 1 , A* : SL= ASC ( A*+CHR* ( O ) >
5SO GOTO 440
590 :
600 input#15,en,en*,et,es
610 if en<20 then return
620 print" {down* disk error"
630 printen;em*;et;es
640 CLOSE l: CLOSE 2: CLOSE 15
Now that you understand how the file pointers work, let's look briefly at how the
various types of files are stored.
13.10 How Sequential Files Are Stored
The directory entry of a sequential file has a file type byte of $81 (129) and the track
and sector pointer points to the first block of the file. A sequential file has a very
simple structure. All the blocks in the file, except the last one, look like this:
Track
Link
Sector
Link
254 bytes o-f the
sequential -file
When we find a block in a sequential file with a track link of zero ($00), we know we
have come to the last block in the file. Normally the second byte in the block would be
the sector link. However, in the last block this byte is a pointer. It points to the last
318
byte in the sector that is part of the file. Any bytes beyond this point are garbage.
Diagrammatically the last block looks like this:
»oo
Link
Pointer to
-final byte
Remaining bytes of
the sequential file
Garbage
13.11 How Program Fifes Are Stored
The directory entry of a program file has a file type byte of $82 (130) and the track
and sector pointer points to the first block of the file. The first block of a program file
is somewhat unusual because it contains the load address for the file. The first block
looks like this:
Track
Link
Sector
Link
Load Address
do byte)
Load Address
(hi byte)
The first 232 bytes
of the program
Let's take another look at the start of the program HOW PART TWO.
. 00: 1 1 'SD 01 04 OA 04 OC 03
Bytes $00 and $01 are the pointer to the next block in the file. In this case they point to
track 17 ($11), sector 13 ($0D).
. oo: li od
OA 04 OC 03
Bytes $02 and $03 are the load address of the file. The load address is in lo byte/hi
byte form. In this case the load address is $0401. If you are more comfortable with
decimal numbers, you can convert to decimal by multiplying the hi byte by 256 and
adding the lo byte like this:
Load Address = 256 x (Value of hi byte) + (Value of lo byte)
= 256x(4) + (l)
= 1024 + 1
= 1025
The remaining 252 bytes in this block are the first 252 bytes of our program.
AH the other blocks in a program file, except the last one, have the same basic
structure as a sequential file.
Track
Link
Sector
Link
254 bytes
of the program
319
When we find a block in a program file with a track link of zero ($00), we know we
have come to the last block in the file. The second byte in the sector points to the last
byte in the sector that is part of the file. Any bytes beyond this point are garbage.
Diagrammatically the lastiblock looks like this:
too
Link
Pointer to
final byte
Remaining bytes
of the program
Garbage
13.12 How User Files Are Stored
A user file does not have any specific form. Apparently Commodore included the user
file type to give programmers a file type that they could use for special purpose files
that were neither program nor sequential files. An example might be a true random
access file. The need for such files was almost entirely eliminated when Commodore
introduced relative files. As a result, very few programs were ever produced that
made use of user files. At present, a user file is used mainly for showmanship and to
intimidate the uninitiated.
If you are planning to create user files, they must have the normal track and sector
links. Otherwise all the blocks in the file will be marked as free in the BAM the first
time you validate the diskette.
13.13 How Relative Files Are Stored
Although relative files are easy to use, their structure is quite complex. A relative file
actually consists of two separate files; a main data file that holds the records and a
separate side-sector file for storing the track and sector numbers of the blocks that
make up the main file.
In order to understand the structure of a relative file, we need one to explore.
Unfortunately, there is no prerecorded relative file on the 1541TEST/DEMO diskette.
We will have to create one. So get your system fired up, we've got some work to do.
INSTRUCTIONS:
If you want your files to be just like the ones shown in the diagrams, follow these
instructions carefully.
1. Start with a diskette that has no files stored on it. A newly formatted diskette will
do just fine.
2. Type in the program listed below.
3. SAVE the program onto the diskette.
4. RUN the program. It will create a simple relative file on your diskette.
*****************
PROGRAM: MAKE REL
*****************
320
100 REM MAKE TEST RELATIVE FILE
110 NR=50:REM NUMBER OF RECORDS IN FILE
120 PRINT" €CLR> MAKING RELATIVE FILE"
130 OPEN 15,8,15, "IO"
140 GOSUB 320ZREM CHECK STATUS
150 OPEN 1,8,5, "TEST REL FILE,L, "+CHR*<2
O)
160 GOSUB 320IREM CHECK STATUS
170 RH=INT<NR/256>:RL=NR-256*RH
180 PR I NT "CDOWN} CREATING FILE OF EMPTY R
ECORDS"
190 PRINT#15,"P"CHR*<5)CHR*(RL)CHR*(RH)C
HR* < 1 )
200 PRINT#1,"LAST RECORD"
210 FOR R=l TO NR
230 : RH=INT(R/256):RL=R-256*RH
240 : PRINT#15, "P"CHR*<5)CHR*(RL)CHR*(RH
) CHR* < 1 >
250 : PRINT"FILLING RECORD #"R
260 : PRINT#1,"TEST RECORD #"R
270 NEXT R
280 PRINT#15, "P"CHR*<5>CHR*(1)CHR*(0)CHR
*<1>
290 CLOSE l: CLOSE 15
300 PRINT"CDOWN>ALL DONE": END
310 :
320 input*15,en.em$,et,es
330 if en<20 or en=50 then return
340 print" {down} disk error"
350 printen;em*;et;es
360 CLOSE l: CLOSE 15: END
Once you have created the test file you may want to use the TRACE FILE program
from Section 13.8 to trace the file. If you followed the instructions, you should see
something like this when you run the TRACE FILE program.
TRACE FILE 1541
FILE NAME? TEST REL FILE
FILE CHAIN
MAIN FILE
17,01 17,12 17,02 17,13 0,239
END OF FILE
SIDE SECTOR FILE
321
17,11 0,23
END OF FILE
ALL DONE
Notice that we seem to have two separate files. A main file that is four blocks long and
a side-sector file that is only one block long.
Let's begin our investigation of how relative files are stored by looking at the directory
entry for our relative file. It should be on track 18, sector 1.
TRACK 18 SECTOR 1
. 00: 00 FF 82 11 00 4D 41 4B MAK
. 08: 45 20 54 45 53 54 20 52 E.TEST.R
. 10: 45 4C AO AO AO 00 00 00 EL
. 18: 00 00 00 00 00 00 03 00
. 20: 00 00 84 11 01 54 45 53 TES
. 28: 54 20 52 45 4C 20 46 49 T.REL.FI
. 30: 4C 45 AO AO AO 11 OB 14 LE
. 38: 00 00 00 00 00 00 05 00
There it is, the second entry in the directory. Let's look at it carefully.
. 20: OO 00 B 11 01 54 45 53 TES
The entry begins with the file type byte (byte $22). The file type is $84 (134) which
indicates that this is a relative file.
. 20: 00 00 84 111 01 54 45 53 TES
Immediately following the file type byte we have the pointer to the first block of the
file. The first block is stored on track 17 ($11), sector 1 ($01). This is the same block
that the TRACE FILE program gives as the first block of the four block main file.
The next bytes of interest are bytes $35 and $36.
. 30: 4C 45 AO AO AO 11 OB 14 LE
The values stored here are 17 ($11) and 11 ($0B). These point to the first block in what
the TRACE FILE program called the side-sector file —track 17, sector 11.
The next byte, $37, is also of interest.
. 30: 4C 45 AO AO AO 11 OB 14 LE
322
The value stored here is 20 ($14). This is the record size we specified when we opened
the relative file.
The last thing we should note is the length of the file.
. 38: 00 00 00 00 00 00 05 00
The file is five blocks long. It seems that the length given in the directory entry is the
sum of the length of the main file (four) and the length of the side-sector file (one).
Now that we know what the directory entry looks like, let's take a look at the two files
themselves. Well start with the main file. It begins on track 17, sector 1.
TRACK 17 SECTOR 1
oo:
11
OC
54
45
53
54
20
52
..TEST.R
08:
45
43
4F
52
44
20
23
20
ECORD.#.
10:
31
20
OD
00
00
00
54
45
18:
53
54
20
52
45
43
4F
52
ST.RECOR
20:
44
20
23
20
32
20
OD
00
D.*.2...
28:
00
00
54
45
53
54
20
52
..TEST.R
30:
45
43
4F
52
44
20
23
20
ECORD.#.
38:
33
20
OD
00
00
00
54
45
3 TE
40:
53
54
20
52
45
43
4F
52
ST.RECOR
48:
44
20
23
20
34
20
OD
00
D.#.4...
50:
00
00
54
45
53
54
20
52
..TEST.R
58:
45
43
4F
52
44
20
23
20
ECORD.tt.
60:
35
20
OD
00
00
00
54
45
5 TE
68:
53
54
20
52
45
43
4F
52
ST.RECOR
70:
44
20
23
20
36
20
OD
00
D.#.6
78:
00
00
54
45
53
54
20
52
..TEST.R
80:
45
43
4F
52
44
20
23
20
ECORD.#.
88:
37
20
OD
00
00
00
54
45
7 TE
90:
53
54
20
52
45
43
4F
52
ST.RECOR
98:
44
20
23
20
38
20
OD
00
Dm fF« Dm • •
AO:
00
00
54
45
53
54
20
52
..TEST.R
A8:
45
43
4F
52
44
20
23
20
ECORD.#.
Bo:
39
20
OD
00
00
00
54
45
9 TE
B8:
53
54
20
52
45
43
4F
52
ST.RECOR
co:
44
20
23
20
31
30
20
OD
D.#.10. .
C8:
00
00
54
45
53
54
20
52
..TEST.R
do:
45
43
4F
52
44
20
23
20
ECORD.tt.
D8:
31
31
20
OD
00
00
54
45
11 TE
eo:
53
54
20
52
45
43
4F
52
ST.RECOR
E8:
44
20
23
20
31
32
20
OD
D.tt.12..
FO:
00
00
54
45
53
54
20
52
..TEST.R
F8:
45
43
4F
52
44
20
23
20
ECORD.tt.
Let's see what we've got.
. 00: 11 0( 54 45 53 54 20 52
TEST.R
323
Bytes $00 and $01 are the usual pointers to the next block in the file. In this case,
track 17 ($11), sector 12 ($0C). Our first record begins at byte $02.
. OO: 11 OC 54 45 53 54 20 52 ..TEST. R
. 08: 45 43 4F 52 44 20 23 20 ECORD.#.
. 10: '31:^1;OD^j60-&/b6l PT4S1 1 TE
. is: 5 3 :; 2-i 52 ±r 43 ^>~ 52 ST.RECOR
. 20: 44 z ;:' 20 32 2c .0 00 d.#.2. ..
The second record appears to start at byte $16. This means that our first record
extends from byte $02 to byte $15 inclusive. A total of 20 characters. This is just what
we specified as the record size when we opened the file. Note that the data we printed
into the record (TEST RECORD # 1) and the carriage return character ($0D) did not
completely fill the record. The data was padded with three null bytes ($00) to fill up
the record.
The structure of the main file is actually quite simple. If you examine the next blocks
(17,12 and 17,02) you'll find they are very similar to this one. Each contains a pointer
to the next block and 254 bytes of data. The only one that is a bit different is the last
block on track 17, sector 13.
00
08
10
18
20
28
30
38
40
48
50
58
60
68
70
78
80
88
90
98
AO
TRACK 17 SECTOR 13
00 EF 53 54 20 52 45 43 ..ST.REC
4F 52 44 20 23 20 33 39 0RD.#.39
20 OD 00 00 54 45 53 54 ....TEST
20 52 45 43 4F 52 44 20 .RECORD.
23 20 34 30 20 OD 00 00 #.40
54 45 53 54 20 52 45 43 TEST.REC
4F 52 44 20 23 20 34 31 0RD.#.41
20 OD OO 00 54 45 53 54 TEST
20 52 45 43 4F 52 44 20 .RECORD.
23 20 34 32 20 OD 00 00 #.42
54 45 53 54 20 52 45 43 TEST.REC
4F 52 44 20 23 20 34 33 0RD.#.43
20 OD 00 00 54 45 53 54 TEST
20 52 45 43 4F 52 44 20 .RECORD.
23 20 34 34 20 OD 00 00 #.44
54 45 53 54 20 52 45 43 TEST.REC
4F 52 44 20 23 20 34 35 0RD.#.45
20 OD 00 00 54 45 53 54 TEST
20 52 45 43 4F 52 44 20 .RECORD.
23 20 34 36 20 OD OO 00 #.46
54 45 53 54 20 52 45 43 TEST.REC
324
. AB: 4F 52 44 20 23 20 34 37 0RD.#.47
. BO: 20 OD 00 OO 54 45 53 54 TEST
. BB: 20 52 45 43 4F 52 44 20 .RECORD.
. CO: 23 20 34 38 20 OD 00 00 #.48
. C8: 54 45 53 54 20 52 45 43 TEST.REC
. DO: 4F 52 44 20 23 20 34 39 0RD.#.49
. D8: 20 OD 00 00 54 45 53 54 TEST
. EO: 20 52 45 43 4F 52 44 20 .RECORD.
. E8: 23 20 35 30 20 OD 00 00 #.50
. FO: FF 00 00 00 00 00 00 00
. F8Z 00 00 OO 00 00 00 00 00
This time the track link is zero ($00) to indicate that this is the last block in the file.
• 00: IB EF 53 54 20 52 45 43 ..ST.REC
Since this is the last block in the file, byte $01 is not a sector link. It is a pointer to the
last byte in the sector that is part of the file.
. 00: 00 EF 53 54 20 52 45 43 ..ST.REC
In this case it has a value of 239 ($EF) which means that the 239th byte is the last byte
in the file.
. D8: 20 OD 00 00 54 45 53 54 TEST
. EO: 20 52 45 43 4F 52 44 20 .RECORD.
. E8: 23 20 35 30 20 OD 00 OO: #. 50. . . .
Note that the 239th byte is the last byte in record number 50.
The structure of the main data file part of a relative file is pretty simple. In fact, it
looks just like a sequential file. The only difference is that each of the records is the
same length.
Now that we have examined the main data file, let's take a look at the side-sector file.
This file is only one block long and is located on track 17, sector 11. Since the side-
sector file is the most complex and misunderstood part of a relative file, well examine
it carefully.
TRACK 17 SECTOR 11
. 00: 00 17 00 14 11 OB 00 00
. 08: 00 00 00 00 00 00 00 00
. 10: 11 01 11 OC 11 02 11 OD
. 18: 00 00 00 00 00 00 00 00 .........
. 20: 00 00 00 00 00 00 00 00
. 28: 00 00 00 00 00 00 00 00
325
30
38
40
48
50
58
60
68
70
78
80
88
90
98
A0
AS
BO
B8
CO
C8
DO
D8
EO
E8
FO
F8
00 00 00 00 00 00 00 OO
00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00
00 00 00 OO OO 00 00 OO
00 00 OO OO 00 00 00 00
00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00
00 00 00 00 OO 00 00 00
00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00
00 00 00 00 OO 00 00 00
00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00
00 00 00 00 00 00 OO 00
It certainly doesn't look like much.
- 00: H^BOO 14 11 OB 00 00 Last block, last byte pointer
If we can go by the usual rules, this is the last block in the file (byte $00 has a value of
zero) and only the first 23 bytes are important (byte $01 has a value of $17).
. 00: 00 17 111 14 11 OB 00 00 .Unused
Byte $02 is a null byte ($00). It is not used.
. 00: 00 17 00 HI 11 OB 00 00 Record size
Byte $03 looks somewhat familiar. It has a value of 20 ($14). It's just the record size of
our file.
. 00: 00 17 00 14 H^SI 00 00 T&SofsidesectorO
Bytes $04 and $05 also seem a bit familiar. They have the values 17 ($11) and 11
($0B). That's the track and sector number for this sector, the first (or zero) block in
the side-sector file. By the way, these are the bytes that the FIX BAD DIR program
326
checks to find side-sector files. If bytes $04 and $05 of the sector are the same as the
track and sector number of the block, we have a potential side-sector file.
. 00: 00 17 00 14 11 OB 00 00 Reserved for T & S numbers
of future side sectors
. 08: 00 00 00 00 00 00 00 00
Bytes $06 to $0F all have the value of zero ($00). They seem to be reserved for
something.
T & S numbers of the blocks
that make up the main file
Bytes $10 to $17 are in use. They have the values 17 ($11) , 1 ($01), 17 ($11), 12 ($0C),
17 ($11), 2 ($02), 17 ($11), and 13 ($0D). Do those numbers seem familiar? They
should. They're the file chain of the main file we traced earlier.
If we expanded our relative file, the file chain stored here would be expanded. Since
we need two bytes to store the track and sector numbers for each block and we have
240 bytes from byte 16 ($10) to 255 ($FF), we have space to store pointers to 240/2 =
120 blocks here.
What happens if we expand our relative file so that we need more than 120 blocks to
hold the data? Well need another block in our side-sector file to hold the rest of the file
chain. When this happens, we'll notice a change in the first part of our old block.
Bytes $00 and $01 will be changed to point to the new block in our side-sector file. In
addition, bytes $06 and $07 will no longer be $00. They will hold the track and sector
numbers of the new block in our side-sector file.
00: 00 17 00 14 11 OB tt ss
T &S of side sector 1
Each block in our side-sector file really stores two file chains. Bytes $04 to $0F hold
the entire file chain of the side-sector file and bytes $10 to $FF all or part of the file
chain of the main data file.
ss #0 ss #1
00: 00 17 00 14
SS #2SS#3 SS#4sg#5 The entire file chain
io: n 01 il^M'^ili^iA l-jJpO
Part or all of the file
;;;;;;;; chain of the main data file
F8: Oc; 00 30 00 00 00 : . oo
327
It only takes 12 bytes to hold the entire file chain of the side-sector file because it is
impossible to produce a side-sector file that is more than six blocks long. Each block
in the side-sector file can hold pointers to 120 blocks. With six blocks in the side-sector
file we can store pointers to 6 x 120 = 720 blocks. Since a diskette only has 664 blocks
free, we never need a seventh sector in the side-sector file. Note that the side sectors
are numbered zero to five.
The Disk Operating System in the 1541 uses the two file chains in the side-sector file
to enable it to locate any record in the file quickly and easily. It works something like
this:
Problem: I have a relative file with a record size of 40. 1 want record number 1206.
How do I locate it?
How many bytes into the file? 1206 x 40 = 48240
How many sectors into the file? 48240/254 = 189.921269 (remember each sector holds
254 data bytes). This is 234 bytes into the 190th sector.
Where is the sector pointer stored? 190/120 = 1.583333
It is the 70th pointer stored in side-sector 1.
Once the location of the pointer has been determined, it is relatively easy for the DOS
to find the pointer, read in the block containing the desired record and position to the
start of the record.
Now that you have a basic understanding of how relative files work, do a bit of
experimenting with them yourself. Expand our test file and see what happens. As
you become more familiar with relative files, you'll find them fast and easy to use.
Just remember, on the 1541 drive you must always specify the byte parameter when
you position to a record and be careful when updating records. See Section 8.6 for
more details.
13.14 An Overview of the Inner Workings of the 1541
Commodore is one of the few computer manufacturers that uses intelligent disk
drives. A 1541 disk drive contains some quite sophisticated electronics. It has a 6502
microprocessor, two 6522 chips to handle input/output operations, 2K of RAM
memory, and a 16K Disk Operating System (DOS 2.6) contained in ROM.
The 6502 microprocessor flips back and forth between being a file manager and a
floppy disk controller. In its file manager mode it:
1. Communicates with the Commodore 64 or VIC-20 computer.
2. Uses and maintains the BAM to keep track of disk storage.
3. Breaks up disk operations into simple read or write jobs.
4. Puts jobs to be done in the job queue.
5. Generates error status reports for the computer.
328
In its floppy disk controller mode the 6502:
1. Monitors the job queue for job requests.
2. Turns the drive motor and drive-active LED on and off.
3. Activates the stepper motor to move the read/write head.
4. Records headers and dummy data blocks to format a diskette.
5. Reads and writes encoded data.
6. Encodes and decodes data.
7. Reports the status of jobs to the file manager.
The memory map below indicates the arrangement of the disk RAM and ROM. It
also highlights a few of the important areas.
SOOOO-HOOll
»0000-*OOFF
♦0100-*01B9
*01BA-*01FF
♦0200-S0229
*022A-*02FF
*0300-*03FF
*0400-*04FF
»0S00-*05FF
♦0600-S06FF
*0700-*07FF
Page
0:
job queue (job & location)
workspace for the 6502
Page
l:
6502 stack
over-flow buffer (encoding)
Page
2:
disk command buffer
workspace for the 6502
Page
3:
data buffer #0
temporary data storage
Page
4:
data buffer ttl
temporary data storage
Page
5:
data buffer #2
temporary data storage
Page
6:
data buffer #3
temporary data storage
Page
3:
data buffer #4
reserved for BAM storage
♦1800-9180F
6522 I/O Chip:
computer/disk drive communication
*1C00-*1C0F
6522 I/O Chip:
floppy disk controller
*Ci00-*F258
♦C100-SF25B
ROM: File manager routines
ROM: Disk controller routines
329
Some of the disk utility programs given in this book bypass the 1541's file manager
and place job requests directly in the job queue. However, a discussion of this
technique is beyond the scope of this book. For more details on this technique and the
internal organization of the 1541 drive, see Inside Commodore DOS, by Richard
Immers and Gerald G. Neufeld, (DATAMOST, 1984).
13.15 The Important Bugs in DOS 2.6
Several bugs have crept into DOS 2.6 and its documentation. The most important of
these are listed below:
1. The byte parameter in the POSITION command is required. It is not optional as
indicated on page 35 of the 1541 User's Manual. It must be present or the drive
will not position to the record correctly. If you leave it off, you will be positioned to
the start of the record only if that record happens to span two sectors. Otherwise
you will be positioned to byte 13 of the record (the DOS assumes the carriage
return character at the end of the command gives the byte position). This
inconsistency has generated much confusion about relative files.
2. The POSITION command can cause problems when revising records. Several
articles in the popular press have indicated that there are serious problems with
the POSITION command on the 2031 and 1541 Commodore disk drives. The
investigations reported in Section 8.6 indicate that the problem is relatively minor
and is easily fixed. Problems occur only when you are writing or revising records
in a non-sequential manner. To fix this bug, simply issue the POSITION command
twice and then create about a 0.5 second pause before using PRINT# to write data
into the record. No fix is needed when reading records or writing records
consecutively (as when filling a file initially). In these situations, the command
appears to be completely reliable.
3. Certain record lengths of relative files are forbidden. The 1541 drive will reject
any command that attempts to open a relative file with a record length of 42 (*), 58
(:) or 63 (?). The reason is that these particular lengths correspond to characters
that have special meanings to the disk operating system.
4. If a replace operation aborts, the old file is lost. If a disk error occurs (including a
70, DISK FULL, error) while you are in the process of replacing a file, the old file
will be lost and the new file will be incomplete. If this happens to you, turn to
Section 11.8 to find out how to recover the old file.
5. The direct-access commands BLOCK-READ and BLOCK-WRITE do not work
correctly. Although these commands are described in the 1541 User's Manual,
they should never be used. Use the command "Ul:" to read a sector and "U2:" to
write a sector instead. The correct syntax for each command is:
SYNTAX:
PRINT*15,"U1: channel* drive* track# sector*"
PRINT#15,"U2: channel* drive# track* sector*"
EXAMPLES:
PRINT*15,"Ul:0 5 18 1" Read track 18, sector 1
330
PRINT#15, "Ul:0 5"TR;SE
PRINT#15, "U2:0 5 18 1" Write track 18, sector 1
PRINT#15,"U2:0 5"TR;SE
For practical examples of the use of "Ul:" and "U2:" commands look through the
listings of the disk utility programs included in this book.
6. The direct-access command BLOCK-ALLOCATE does not always work correctly.
This command is used to allocate a particular sector in the BAM. If the specified
sector is free, the command will work properly. However, if the specified sector is
already allocated, an entire trackful of sectors will be allocated instead. This is
why the UNSCRATCHER program reads in the BAM and checks to be sure that
all the blocks are free before the blocks in the recovered file are allocated using a
BLOCK-ALLOCATE command. In case you are wondering why you would ever
want to allocate an already allocated sector, the reason is that this technique was
used for creating random access files in the days before relative files. When you
attempt to allocate an already allocated block, the drive is supposed to report the
track and sector numbers of the next available block as part of the 65, NO BLOCK,
error message. Not only does the drive preallocate a track full of sectors, it will
even report sectors on track 18 (the reserved directory track) as the next available
block.
7. The drive reset command "U" does not work. According to page 39 of the 1541
User's Manual, you may use either PRINT#15,"U" or PRINT#15,"U;" to reset the
drive to its power-on status. If you use the PRINT#15,"U" command, the drive will
lock up. You will have to turn the drive OFF and ON again to recover. To reset the
drive you must use the "U;", "U9", or "UF commands.
8. The drive slow-down command "U+" does not work. According to page 39 of the
1541 User's Manual, you can use the command PRINT#15,"U+" to set the drive's
communication speed so it is compatible with the Commodore 64. If you issue this
command, the drive will lock up. You will have to turn the drive OFF and ON
again to recover. However, don't worry. As you have undoubtably discovered by
now, your Commodore 64 and your 1541 communicate quite well without the use
of this command.
Conclusion
Although the 1541 is somewhat slow and has a few flaws, it is a very sophisticated
disk drive. It's low price makes it an outstanding bargain. None of the other drives in
its price range provide relative file handling and make use of asynchronous I/O or
read-ahead buffering. Commodore is to be congratulated for a fine product.
331
APPENDIX
A
GLOSSARY
allocated sector— a sector (block) that is identified as being currently in use in the
diskette's Block Availability Map (BAM). When a program or data file is recorded on
a diskette, the status of sectors used to store the file are automatically changed from
unallocated (available for use or free) to allocated (currently in use) by the 1541's Disk
Operating System (DOS).
application program— a program that carries out a useful, real world task.
Examples include word processors, mailing list programs, grade book programs and
electronic spreadsheets.
BAM— an acronym for Block Availability Map.
binary chop— a technique for rapidly scanning a sorted list of numbers or character
strings. The technique involves checking if the desired value is in the first or last half
of the list. Once this is determined, the process is repeated using only the half of the
list that contains the desired value. Each check splits the remaining list in half.
Hence the term binary (in two) chop (to split).
block— a synonym for sector. It can store 256 characters of information.
Block Availability Map— a map which indicates which sectors (blocks) on a diskette
are currently being used (in use or allocated) to store part of a program or data and
which sectors are still available for use (free or unallocated). It is often referred to as
the BAM. The status of each sector is represented by a single bit. If the bit cor-
responding to a particular sector (block) is one, the sector is unallocated. If the bit is
zero, that sector has been allocated. The BAM is stored on track 18, sector 0. When a
diskette is initialized, a copy of the BAM is read into the disk drive's RAM memory
and resides at memory locations $0704-$078F.
buffer— an area of RAM memory inside the 1541 disk drive that is used to tem-
porarily hold data that is being read from, or written to, a diskette. When data to be
recorded on diskette is sent from the computer to the disk drive, it is stored in one of
the buffers until 256 characters (bytes) of data have been received. Then all 256 bytes
are encoded and recorded as the data block of one sector on the diskette. When data is
to be read from diskette, the 256 bytes that make up the data block of a particular
sector are read, decoded and stored in a buffer. As the data is requested by the
333
computer, it is sent from the buffer. When all 256 bytes have been sent, another
buffer full of bytes is read from the diskette. The 1541 disk drive has five buffers (#0
to #4).
close a file— when you are finished reading from or writing to a file on a diskette, the
file should be closed. This breaks the communication link that was established when
the file was first opened for access. It is particularly important to close a file that was
opened in write mode. Unless the file is closed, the last part of the file will not be
recorded onto the diskette. It will simply remain in the buffer inside the 1541. In
addition, the directory entry for that file will not be completed and you will end up
with an unclosed file. Unclosed files have an asterisk (*) beside their file types in a
directory listing (*PRG, *SEQ or *USR).
channel number (channel*)— an integer between zero and 15. It is used in an
OPEN statement to identify the type of communication link desired between the
computer and the disk drive. Channel* 0 is reserved for use in loading programs.
Channel* 1 is reserved for use in saving programs. Channel's 2 through 14 are for
general use in reading or writing any type of file. Channel* 15 is the command
channel and is used to send housekeeping commands to the disk drive or to read the
disk drive's error status report. The channel number is also known as the secondary
address.
CHR$— a BASIC function that allows you to convert the ASCII code for a character
to the actual character itself. For example, the ASCII code for the letter A is 65. To
print an A on the screen you could say either PRINT "A" or PRINT CHR$(65).
command channel— one type of communication channel that can be established
between your computer and the 1541 disk drive. It is used for sending disk house-
keeping commands to the disk drive and for reading the disk drive's error status. The
command channel is channel* 15.
decimal notation— refers to numbers written in base 10. Numbers that we encounter
in normal, everyday life are written in base 10. If you are an incurable hacker and
have difficulty in expressing numbers in anything other than binary or hexadecimal
notation, tune in to Sesame Street.
delimiter— a character that is used to separate the various parts (fields) of a record in
a data file. For example, a delimiter might be used to separate a person's name (the
name field) from his or her address (the address field). A delimiter can be any
character that would never appear in either of the fields to be separated. Usually a
carriage return character, CHR$(13), or a comma (,), is used as a delimiter because
these characters are delimiters recognized by the INPUT statement and naturally
divide the data into two fields.
device number (device*)— an integer that is used to identify a particular device. On
the VIC-20 and Commodore 64, devices with numbers greater than four are assumed
to be connected to computer by the serial bus. The 1541 comes from the factory as
device* 8. It may be changed to device* 9, 10, 11 or 12, either temporarily or
permanently.
334
directory — a special purpose file stored on track 18 of a diskette. It contains the
diskette's Block Availability Map and a list of file entries. There is one entry for each
file stored on the diskette. Each entry holds a file's name, type, size, and the track and
sector numbers of the sector used to store the first block of that file. You can OPEN
the directory file for read access. It is a program file and its file name is $. If you use a
data channel# (2-14), you get to read the whole file. If you use channel* 0, you read the
pseudo BASIC directory that is loaded by LOAD "$",8.
directory entry— refers to all the information stored in the diskette's directory about
a particular file. The entry consists of the file type, the track and sector number of the
first block in the file, the file name, and the length of the file (in blocks). An entry for a
relative file also contains the record size, and the track and sector number of the first
block in the side-sector file.
directory file — see directory.
disk— synonym for floppy diskette.
disk command— a command sent from the computer to the 1541 disk drive instruct-
ing it to carry out a specific disk operation. The 1541 commands described in detail in
this book are NEW, INITIALIZE, SCRATCH, RENAME, COPY, VALIDATE and
POSITION. The 1541 command set also includes "Ul:" (read a sector), "U2:" (write a
sector), "M-R" (memory read), "M-W" (memory write), "M-E" (memory execute),
"B-A" (block allocate), "B-F" (block free), "B-E" (block execute), "U0" (reset jump
vectors), "UI-" (speed up data transmission for the VIC-20), and "&" (load and execute
a utility). Some of these more advanced commands are used in the disk utility pro-
grams. If you are interested in how to use these commands more effectively, see the
more advanced book, Inside Commodore DOS, by Richard Immers and Gerald G.
Neufeld, (DATAMOST, 1984).
disk error status— the 1541's Disk Operating System (DOS) monitors all disk opera-
tions. When an operation has been completed or aborted because of an error, the DOS
generates a status report. The report consists of an error code number, an English
language error message, and the track and sector number of where the error, if any,
occurred. The status report is read using the command channel.
disk ID— a two character code used by the Disk Operating System to identify a
diskette. The disk ID is specified when the diskette is formatted (newed). A diskette's
ID cannot be changed unless the entire diskette is reformatted. It is good practice to
have a different disk ID for every diskette you have.
disk name— an arbitrary name used to help you identify a diskette. The name is
specified when the diskette is formatted. It may be changed using the MOD DISK
NAME program (Appendix E).
Disk Operating System— a machine language program that directly controls the
operation of the disk drive. It is often called the DOS. The 1541 uses Commodore DOS
version 2.6. It is permanently stored in 16K of Read Only Memory (ROM) in the disk
drive.
335
disk utility— a program that performs some useful task related to the preservation of
the sanity of disk drive owners. Examples of disk utilities include the
UNSCRATCHER, which recovers a scratched file, CHECK DISK, which checks a
diskette for flaws, and CONFIRM ALL FILE, which checks to see if a diskette's
BAM and directory are correct and that all files are intact
diskette— a circular disk of thin plastic, coated on both sides with a magnetic material
and encased in a protective jacket. It is used for storing computer programs or data
files.
DOS— an acronym for Disk Operating System.
DOS 5.1— this is not a Disk Operating System. It is a short, machine language
program that makes it easier to view a diskette's directory, load or save programs,
send disk commands, and read the disk drive's error status. This program resides at
the top of memory and does not affect the normal operation of BASIC programs in
the VIC-20 or Commodore 64.
double density— describes a diskette that is designed to be used with a disk drive
that uses double density recording techniques. In double density recording the data is
encoded so that clock bits are not needed between data bits. The 1541 uses GCR
encoding to do this. Since there are no clock bits, a double density diskette is rated at
5,876 flux changes per inch and 5,876 bits per inch along a track. Note that both
single and double density diskettes can record 5,876 flux changes per inch. Since the
1541 records a maximum of 5,800 flux changes per inch, you can use either single or
double density diskettes with a 1541 drive. The reason that double density diskettes
are usually more expensive is that they are also designed to work with drives that
write up to 80 tracks per inch. Since the 1541 only uses 40 tracks per inch, single
density diskettes are adequate.
double-sided— describes a diskette that is designed to be used with a disk drive that
can record on both sides of the diskette. Since the 1541 only records on one side of the
diskette, you can use the less expensive, so-called single-sided diskettes.
drive-active LED— the red, Light Emitting Diode (LED) on the front of the 1541
disk drive. If this light is on, the drive is busy carrying out some disk operation. If it is
flashing on and off regularly, a disk error has occurred. It is normal for the light to
blink on and off occasionally as you load a long program. Normally, you should never
open the drive door while the light is glowing steadily. However, if you realize you
have just told the drive to NEW your only copy of a valuable diskette, quickly open
the door and rescue that diskette!
end of file— when reading a file, the 1541 sends a special end-or-identify (EOI) signal
to your computer along with the very last character from the file. When your computer
receives this EOI signal, it sets the status variable (ST) to 64. If you check the value
stored in ST after every INPUT# or GET# statement, you can tell when you read the
last piece of data in your file.
file identification information— the information included in an OPEN statement
that tells the disk drive the name and type of the file you want to access. For sequential,
program and user files you must also specify the mode of access desired (read, write
or modify). This is not necessary for relative files because these files are always
opened for read/write access. However, if the OPEN statement will create a new
relative file, you must specify the record length (size).
file type— identifies a file as being of a particular type. The 1541 DOS supports four
different types of files: program files (PRG), sequential files (SEQ), relative files
(REL), and user files (USR). An asterisk (*) in front of the file type in a directory
listing identifies an unclosed file (e.g., *SEQ, *PRG, *USR). A less than sign (<)
following the file type identifies a locked file (e.g., SEQ<, PRGK, REL< or USR<). A
locked file cannot be scratched unless it is first unlocked.
file type byte— the single character (byte) in a file's directory entry on track 18 that
identifies the type of file. Normal values are 128 ($80), a DEL file; 129 ($81), a SEQ
file; 130 ($82), a PRG file; 131 ($83), a USR file; and 132 ($84), a REL file. Changing
the file type byte from $8x to $Cx locks the file.
file number (f ile#) — an integer between one and 127 given in an OPEN statement to
identify that particular file. Subsequent GET#, INPUT#, PRINT# and CLOSE state-
ments that refer to this file must include the file#. An identifying number is needed
because the VIC-20 and Commodore 64 allow the user to have up to 10 different files
open simultaneously. Once a file has been closed the f ile# that was used to refer to that
file has no further significance. The file# is also referred to as the logical file number.
flippy— a diskette that has been modified by cutting a second write protect notch in
the protective jacket. This allows the second side of the diskette to be used for storing
programs and data when the diskette is inserted into the drive upside down.
free sector— a sector on a diskette that is not currently being used to store part of a
program or data file. It is also known as an unallocated sector. The diskette's Block
Availability Map (BAM) is used to keep track of which sectors are free and which
have been allocated (are in use).
format a diskette— recording a pattern of information on a new, blank diskette so
that it can be used for storing programs or data. This is also known as newing a
diskette. During the formatting process, an identifying header block and a dummy
data block are recorded for each of the 683 sectors on the diskette. Any information
previously recorded on the diskette is irretrievably lost. On the 1541 formatting a
diskette takes approximately 80 seconds.
GCR— an acronym for Group Code Recording. Each 8-bit data byte is encoded into a
10-bit GCR equivalent before being recorded onto the diskette surface. The data is
decoded into 8-bit form when it is read back from the diskette.
hex— short for hexadecimal notation (e.g., $0A = 10).
hexadecimal notation— refers to numbers written in base 16 rather than our more
familiar base 10 (decimal) notation. A number written in hexadecimal notation is
usually preceded by a dollar sign ($4F). Hexadecimal notation is used by most
computer buffs because it allows all numbers between zero and 255 to be expressed
as two digits ($00-$FF) and it simplifies the conversion of numbers to binary (base 2)
notation. Appendix B contains a table giving all the hexadecimal numbers between
$00 and $FF and their decimal equivalents.
337
I/O— an abbreviation for input/output.
I/O status— refers to whether an input or output operation has been completed suc-
cessfully or not. In the VIC-20 and Commodore 64 the I/O status is automatically
stored in the status variable, ST. If the value stored in ST is zero, the operation has
been completed satisfactorily. A value of 64 after a read operation indicates that you
have reached the end of the file.
index file— the file that serves as an index into another file. See indexed relative file.
indexed relative file— a way of storing and retrieving information by putting
unsorted data in a large relative file and using the data in another file as an index to
the relative file. The index allows the data in the relative file to be retrieved or
scanned quickly in a sorted order.
initializing a diskette— reading a copy of the Block Availability Map (BAM) recorded
on track 18, sector 0 of the diskette into the disk drive's RAM memory. This ensures
that the disk drive recognizes the diskette properly.
link— the pointer that gives the track and sector numbers of the sector (block) on the
diskette where the next part of a file is stored. For more information see Section 13.9.
LOAD— a BASIC command that reads a program that is stored on a cassette tape or
diskette and stores a copy of this program in the computer's memory. The copy of the
program on tape or diskette is not erased or modified in any way.
loader program— a program that is designed to load another program or other
information into memory and run it. Many commercial programs use loader programs
that check to make sure you have an original copy of the tape or diskette before they
will load and run the main program. Usually these loader programs are written in
machine code and make use of obscure programming tricks, encoding and illegal
instructions to make it very difficult for someone to determine exactly what the
loader does.
machine language— the native language of the 6502/6510 microprocessor in your
computer and disk drive. Each line in a BASIC program must be interpreted and
converted into a set of equivalent machine language instructions before the micro-
processor can carry out the requested operation. Programs written in machine
language run up to 1,000 times faster than an equivalent BASIC program.
machine language monitor— a program that makes it easy to examine and change
the contents of various memory locations in your computer. Some machine language
monitor programs include a simple assembler to make programming in machine
language easier and a simple disassembler to convert machine language programs
into more readable form. SUPERMON on the C-64 DISK PACK is an example of a
machine language monitor.
monitor— a TV set without commercials.
opening a file— means establishing a communication link between your computer
and some device so that you can output information to the device or input information
from the device. A useful analogy is opening the drawer of a filing cabinet so that you
can put information into it or take information out of it.
position to a record— a disk command that tells the Disk Operating System (DOS)
that you want to access a particular part (record) of a relative file. You must specify
the channel# you used when you opened the file and the number of the record you
want. The DOS calculates where that particular record is stored on the diskette and
reads it into the disk drive's RAM memory.
program— a set of instructions that tells the computer exactly how to do something
(e.g., play a game). Programs can be written in the native language of the computer's
microprocessor (machine language) or a language that is more understandable to
humans like BASIC.
RAM— an acronym for Random Access Memory. A place where the microprocessor
can store information (such as a program) and read it back later. The information is
not stored permanently. When new information is stored in a particular location, any
existing information stored there is lost. When the power is turned off, all the infor-
mation stored in RAM memory is lost.
random access file— a data file in which any data item (record) can be read or
written without having to read or write the entire file. A relative file is a random
access file that has a fixed record size.
read error— occurs when the disk drive has attempted to find a particular sector on
the diskette and read its data block, and found that the information recorded on
diskette for this sector was incorrect in some way. An error in either the identifying
header block or the data block will cause a read error.
reading a file— the disk drive plays back (reads) all or part of a group of related
information (file) that has been previously recorded on a diskette.
record— the part of a file that contains all the information stored about a particular
person, place or thing. In a sequential file, the length of the records usually varies
according to the amount of information stored about that particular person, place or
thing. In a relative file, the records are all the same length.
relative file— Commodore's name for a random access data file. All the records in a
relative file must be the same length. Individual records are identified by number.
When a particular record is to be accessed, the record number is specified in a
POSITION command. The disk operating system accesses the appropriate sector(s)
on the diskette and the record may be read or written.
replace— when an "at" sign and a colon precede the file name in a SAVE or OPEN
command (e.g., SAVE "@:BIG TEST",8), the new program or data file will replace
the older version stored on the diskette and use the same name. The new file is saved
before the older version is replaced. If an error causes the writing of the new file to be
aborted, the old file is normally lost. If the old file was normal, the old directory entry
is used. If the old file was unclosed, a new directory entry is created and the old file
type byte changed to $00 (scratched).
339
ROM— an acronym for Read Only Memory. The place where the instructions that
control the basic operation of your computer or disk drive are stored. These instruc-
tions may not be changed without replacing the memory chips. This information is
not erased when the power is turned off.
SAVE— a BASIC instruction that records a copy of the program currently in the
computer's RAM memory onto cassette tape or diskette.
scratch a file— to delete an unwanted file from the directory of a diskette. The file
itself is not erased and may be recovered if no new information has been recorded on
the diskette since the file was scratched.
secondary address— a synonym for channel*.
sector— a segment of a circular ring (a track) on the surface of a diskette which
contains a block of information that identifies which sector it is (the header block) and
includes a block of 256 characters (bytes) of data (the data block). Sectors are identified
by their track number and their sector number. The header block and a dummy data
block are recorded on the diskette when a diskette is formatted.
sector number— a number which identifies a particular sector on a track. The sectors
that make up a track are numbered consecutively from zero to the maximum for that
track. Tracks 1 to 17 have 21 sectors (0 to 20), tracks 18 to 24 have 19 sectors (0 to 18),
tracks 25 to 30 have 18 sectors (0 to 17), and tracks 31 to 35 have 17 sectors (0 to 16).
sequential file— a file of information in which the individual pieces of information
(records) are packed one behind the other. The size of an individual record varies with
the amount of information stored about that particular person, place or thing. To read
a particular record in a sequential file you must read through ^11 the preceding
records in the file. A record cannot be modified without rewriting the entire file.
serial bus— the cable that connects the 1541 disk drive to the Commodore 64 or
VIC-20 computer. Information that passes through this bundle of wires (a bus)
between the two devices is transmitted one bit at a time (serially), hence the name.
side sector— one of the sectors that is used by the disk operating system of the 1541 to
store the track and sector numbers of the blocks that make up a relative file. Each
side sector stores the record size, the track and sector numbers of all other side
sectors, and the track and sector numbers of up to 120 blocks on which the main data
file is stored. Since the maximum length of any file on a 1541 formatted diskette is
664 blocks, there are a maximum of six side sectors needed to hold the pointers for a
relative file.
side-sector file— a short auxiliary file set up and maintained by the disk operating
system of the 1541 to keep track of the track and sector numbers of the blocks on
which the main data in a relative file is stored.
single density— describes a diskette that is designed to be used with a disk drive that
uses single density recording. A single density diskette is designed to record up to
5,876 flux changes per inch along a track. Since single density recording records a
340
clock bit between each data bit, up to 2,938 data bits can be recorded per inch. The
1541 drive does not use normal single density recording. It uses GCR encoding which
eliminates the need for the clock pulses. Data is recorded at 4,100 to 5,800 bits per
inch. Since there is at most one flux change per bit, there are a maximum of 5,800
flux changes per inch along a track. As a result, the 1541 can use single density
diskettes rated at 40 tracks per inch.
single-sided— describes a diskette that is designed to be used with a disk drive that
records on only one side of the diskette (the back). Since the 1541 only records on one
side of the diskette, this is the kind of diskette to purchase for your 1541.
status variable— a special variable, ST, in the VIC-20 and Commodore 64 computers
that is automatically set by the computer to indicate the status of the most recent
input or output operation. If ST=0, the most recent operation was completed satis-
factorily. If ST=64, you have just read the last item of data in the file being read.
track— a circular ring on the surface of a diskette that is used for recording infor-
mation. On a 1541 formatted diskette there are 35 concentric tracks numbered 1 to
35. Track 1 is the outermost track, closest to the outer edge of the disk. Track 35 is the
innermost track, closest to the center of the disk.
track and sector link— the track and sector numbers of the sector that stores the
next block of a file. The directory entry contains the link to the first sector in the file;
the first two bytes stored in the first sector are the link to the second sector; the first
two bytes stored in the second sector are the link to the third sector, and so on....
unallocated sector— a sector (block) that is identified as being available for use (free)
in the diskette's Block Availability Map (BAM). When a program or data file is
scratched, the status of sectors used to store the file is automatically changed from
allocated (currently in use) to unallocated (free or available for use) by the 1541's disk
operating system.
validate a diskette— creating a new copy of a diskette's Block Availability Map
(BAM) by tracing through all the active files listed in the directory and marking
those sectors used to store these files as allocated (currently in use). All other sectors
on the diskette are marked as unallocated (free or available for use).
verify— comparing the copy of a program stored in the computer's memory with the
copy stored on cassette tape or diskette. If the copies match exactly, the computer
reports OK. If there is any difference between the copies, the computer reports
?VERIFY ERROR. If this happens to you after saving a program, remove the
diskette, turn the drive OFF and ON, insert a different diskette and try the SAVE/
VERIFY sequence again.
wedge— a general term describing a machine language routine that modifies the
part of BASIC that interprets commands so that the new routine can check for
special characters or commands. Most routines of this kind are designed to add new
commands to BASIC. These new commands are said to be wedged into BASIC. The
DOS 5.1 routine does this.
341
write protect notch— the % inch square notch cut out of one edge of the jacket of a
floppy diskette. The presence or absence of this notch is detected by the disk drive. If
the notch is present, you can record information on the diskette. If there is no notch, as
on some commercially recorded diskettes, or if the notch is covered by an opaque tab,
recording is disabled.
write protect tab— a small gummed tab designed to be folded over the edge of the
jacket of a floppy diskette to cover the write protect notch. If the tab is in place, you
cannot record any new information on the diskette.
writing a file— the procedure carried out by the disk operating system in response to
OPEN, PRINT# and CLOSE commands sent by the computer, in which the DOS
creates a new directory entry, records the new information on the diskette, and
updates the diskette's Block Availability Map to indicate that the sectors used to store
the information are now in use (allocated).
342
APPENDIX
B
REFERENCE TABLES
Hexadecimal/Decimal Equivalents
HEX
DEC
HEX
DEC
HEX
DEC
HEX
DEC
HEX
DEC
HEX DEC
HEX
DEC
HEX DEC
$00
0
$20
32
$40
64
$60
96
$80
128
$A0 160
$C0
192
$E0 224
$01
1
$21
33
$41
65
$61
97
$81
129
$A1 161
$C1
193
$E1 225
$02
2
$22
34
$42
66
$62
98
$82
130
$A2 162
$C2
194
$E2 226
$03
3
$23
35
$43
67
$63
99
$83
131
$A3 163
$C3
195
$E3 227
$04
4
$24
36
$44
68
$64
100
$84
132
$A4 164
$C4
196
$E4 228
$05
5
$25
37
$45
69
$65
101
$85
133
$A5 165
$C5
197
$E5 229
$06
6
$26
38
$46
70
$66
102
$86
134
$A6 166
$C6
198
$E6 230
$07
7
$27
39
$47
71
$67
103
$87
135
$A7 167
$C7
199
$E7 231
$08
8
$28
40
$48
72
$68
104
$88
136
$A8 168
$C8
200
$E8 232
$09
9
$29
41
$49
73
$69
105
$89
137
$A9 169
$C9
201
$E9 233
$0A
10
$2A
42
$4A
74
$6A
106
$8A
138
$AA 170
$CA
202
$EA 234
$0B
11
$2B
43
$4B
75
$6B
107
$8B
139
$AB 171
$CB
203
$EB 235
$0C
12
$2C
44
$4C
76
$6C
108
$8C
140
$AC 172
$CC
204
$EC 236
$0D
13
$2D
45
$4D
77
$6D
109
$8D
141
SAD 173
$CD
205
$ED 237
$0E
14
$2E
46
$4E
78
$6E
110
$8E
142
$AE 174
$CE
206
$EE 238
$0F
15
$2F
47
$4F
79
$6F
111
$8F
143
$AF 175
$CF
207
$EF 239
$10
16
$30
48
$50
80
$70
112
$90
144
$B0 176
$D0
208
$F0 240
$11
17
$31
49
$51
81
$71
113
$91
145
$B1 177
$D1
209
$F1 241
$12
18
$32
50
$52
82
$72
114
$92
146
$B2 178
$D2
210
$F2 242
$13
19
$33
51
$53
83
$73
115
$93
147
$B3 179
$D3
211
$F3 243
$14
20
$34
52
$54
84
$74
116
$94
148
$B4 180
$D4
212
$F4 244
$15
21
$35
53
$55
85
$75
117
$95
149
$B5 181
$D5
213
$F5 245
$16
22
$36
54
$56
86
$76
118
$96
150
$B6 182
$D6
214
$F6 246
$17
23
$37
55
$57
87
$77
119
$97
151
$B7 183
$D7
215
$F7 247
$18
24
$38
56
$58
88
$78
120
$98
152
$B8 184
$D8
216
$F8 248
$19
25
$39
57
$59
89
$79
121
$99
153
$B9 185
$D9
217
$F9 249
$1A
26
$3A
58
$5A
90
$7A
122
$9A
154
$BA 186
$DA
218
$FA 250
$1B
27
$3B
59
$5B
91
$7B
123
$9B
155
$BB 187
$DB
219
$FB 251
$1C
28
$3C
60
$5C
92
$7C
124
$9C
156
$BC 188
$DC
220
$FC 252
$1D
29
$3D
61
$5D
93
$7D
125
$9D
157
$BD 189
$DD
221
$FD 253
$1E
30
$3E
62
$5E
94
$7E
126
$9E
158
$BE 191
$DE
222
$FE 254
$1F
31
$3F
63
$5F
95
$7F
127
$9F
159
$BF 192
$DF
223
$FF 255
343
Commodore 64/VIC-20 Internal Storage
of BASIC Characters and Tokens
HEX
DEC
CHR
HEX
DEC
CHR
HEX
DEC TOKEN
$20
32
SPC
$40
64
@
$80
128 END
$21
33
i
$41
65
A
$81
129 FOR
$22
34
it
$42
66
B
$82
130 NEXT
$23
35
#
$43
67
C
$83
131 DATA
$24
36
$
$44
68
D
$84
132 INPUT*
$25
37
%
$45
69
E
$85
133 INPUT
$26
38
&
$46
70
F
$86
134 DIM
$27
39
1
$47
71
G
$87
135 READ
$28
40
(
$48
72
H
$88
136 LET
$29
41
)
$49
73
I
$89
137 GOTO
$2A
42
*
$4A
74
J
$8A
138 RUN
$2B
43
+
$4B
75
K
$8B
139 IF
$2C
44
i
$4C
76
L
$8C
140 RESTORE
$2D
45
-
$4D
77
M
$8D
141 GOSUB
$2E
46
.
$4E
78
N
$8E
142 RETURN
$2F
47
/
$4F
79
0
$8F
143 REM
$30
48
0
$50
80
P
$90
144 STOP
$31
49
1
$51
81
Q
$91
145 ON
$32
50
2
$52
82
R
$92
146 WAIT
$33
51
3
$53
83
S
$93
147 LOAD
$34
52
4
$54
84
T
$94
148 SAVE
$35
53
5
$55
85
U
$95
149 VERIFY
$36
54
6
$56
86
V
$96
150 DEF
$37
55
7
$57
87
W
$97
151 POKE
$38
56
8
$58
88
X
$98
152 PRINT#
$39
57
9
$59
89
Y
$99
153 PRINT
$3A
58
:
$5A
90
Z
$9A
154 CONT
$3B
59
#
$5B
91
[
$9B
155 LIST
$3C
60
<
$5C
92
£
$9C
156 CLR
$3D
61
=
$5D
93
]
$9D
157 CMD
$3E
62
>
$5E
94
1
$9E
158 SYS
$3F
63
j
$5F
95
<-
$9F
159 OPEN
HEX DEC TOKEN
$A0
160
CLOSE
$A1
161
GET
$A2
162
NEW
$A3
163
TAB(
$A4
164
TO
$A5
165
FN
$A6
166
SPC(
$A7
167
THEN
$A8
168
NOT
$A9
169
STEP
$AA
170
+
SAB
171
-
$AC
172
*
$AD
173
/
$AE
174
1
$AF
175
AND
$B0 176 OR
$B1 177 >
$B2 178 =
$B3 179 <
$B4 180 SGN
$B5 181 INT
$B6 182 ABS
$B7 183 USR
184 FRE
185 POS
$BA 186 SQR
$BB 187 RND
$BC 188 LOG
$BD 189 EXP
$BE 191 COS
$BF 192 SIN
$B8
$B9
HEX
DEC
TOKEN
$C0
192
TAN
$C1
193
ATN
$C2
194
PEEK
$C3
195
LEN
$C4
196
STR$
$C5
197
VAL
$C6
198
ASC
$C7
199
CHR$
$C8
200
LEFT$
$C9
201
RIGHT$
$CA
202
MID$
$CB
203
GO
$CC
204
Values
$CD
205
from
$CE
206
$CC
$CF
207
to
$D0
208
$DA
$D1
209
are
$D2
210
tokens
$D3
211
only
$D4
212
in
$D5
213
BASIC 4
$D6
214
and are
$D7
215
unused
$D8
216
on the
$D9
217
VIC-20
$DA
218
& C-64
$DB
219
Values from $DC
to $FF are not
in BASIC
2 & 4.
NOTES:
1. The hexadecimal and decimal numbers given in this table are the code numbers
used internally by the Commodore 64 and VIC-20 computers to represent the
character shown to the right of the pair of numbers. When you type in a program
you type characters and words such as A and PRINT. The computer's screen
editor takes these characters and converts them into ASCII codes (e.g., A = 65) and
BASIC tokens (e.g., PRINT = 153). It is only when you LIST a program that these
internal codes are converted back to letters and words.
2. $00 (zero) is the last byte in any BASIC line.
3- Jo?i™ KiS included ^ a REM statement, a LIST of the program will stop with a
?SYNTAX ERROR when this line is reached.
4. A line that contains characters other than those listed above may LIST but will
halt with a ?SYNTAX ERROR during a RUN.
5. Any character may appear inside a BASIC statement provided it is preceded bv a
quotation mark, CHR$(34). y
344
Number of Sectors per Track on the 1541
Zone
#1
#2
#3
#4
Tracks
Sectors
Number
Numbered
Numbered
of Sectors
ltol7
0to20
21
18 to 24
0tol8
19
25 to 30
0tol7
18
31 to 35
0tol6
17
Location of Important Parts of a Directory Sector
Entry Number within a Directory Sector
#0 #1 #2 #3 #4 #5 #6 #7
File type byte
$02
$22
$42
$62
$82
$A2
$C2
$E2
Start block (t#)
$03
$23
$43
$63
$83
$A3
$C3
$E3
Start block (s#)
$04
$24
$44
$64
$84
$A4
$C4
$E4
File name
$05-14
$25-34
$45-44
$65-74
$85-94
$A5-B4
$C5-D4
$E5-F4
Side sector (t#)
$15
$35
$55
$75
$95
$B5
$D5
$F5
Side sector (s#)
$16
$36
$56
$76
$96
$B6
$D6
$F6
Record size
$17
$37
$57
$77
$97
$B7
$D7
$F7
Length (lo byte)
$1E
$3E
$5E
$7E
$9E
$BE
$DE
$FE
Length (hi byte)
$1F
$3F
$5F
$7F
$9F
$BF
$DF
$FF
Normal Filling Sequence of the Directory Track
(8 entries/sector)
First 48 entries
18/1 ->
18/4 ->
18/7 ->
18/10 -> 18/13 -> 18/16
Next 48 entries
18/2 ->
18/5 ->
18/8 ->
18/11 -> 18/14 -> 18/17
Last 48 entries
18/3 ->
18/6 ->
18/9 ->
18/12 ~> 18/15 -> 18/18
File Type Bytes and Their Meaning
DEL
SEQ
PRG
USR
REL
Unclosed file
$00 (0)
$01 (1)
$02 (2)
$03 (3)
$04 (4)
Normal file
$80 (128)
$81 (129)
$82 (130)
$83 (131)
$84 (132)
During replace
$A0 (160)
$A1 (161)
$A2 (162)
$A3 (163)
$A4 (164)
Locked file
$C0 (192)
$C1 (193)
$C2 (194)
$C3(195)
$C4 (196)
345
APPENDIX
C
DISK ERROR MESSAGES
All error messages have the following general form:
SYNTAX:
error*, message, track#, sector*
EXAMPLES:
00, OK, 00, 00
21, READ ERROR, 22, 05
25, WRITE ERROR, 12,15
31, SYNTAX ERROR, 00, 00
WHERE:
error# = the error code number which indicates the exact type of
error,
message = the English language error message that indicates the
general class of error.
track# = the track number of the sector where the error, if any,
occurred.
sector# = the sector number of the sector where the error, if any,
occurred.
NOTE:
Some of the errors are WRITE errors while others are READ errors.
00: OK
Disk operation completed satisfactorily.
01: FILES SCRATCHED
Informational message produced after scratching one or more files. In this case the
number following the English language message is the number of files that has been
scratched. Since this is not an error, the drive-active LED will not be flashing. This
message may be ignored.
347
20: READ ERROR
When a particular sector is to be read, the disk controller creates a search image of
the header block. It then attempts to find a sector header on the track that matches
the image. After examining 90 headers without success, the disk controller gives up
and this error is reported. This error is usually caused by switching the disk drive on
or off with a diskette in it. It may also be part of a disk protection scheme.
21: READ ERROR
While searching for the desired sector header, the disk controller is unable to find a
SYNC mark on the track within the allowed time. This error is usually caused by
attempting to read an unformatted diskette. It may also be part of a disk protection
scheme.
22: READ ERROR
The disk controller has found the desired header block. However, when attempting to
read the data block, it found that the first character in the data block was not a $07
character (the data block ID character). This error is usually caused by an alignment
problem or swapping diskettes between 4040 and 1541 disk drives. It may also be
part of a disk protection scheme.
23: READ ERROR
The disk controller has found the desired header block and read in the data block.
However, when checking the data read, the computed checksum did not match the
checksum that was recorded on the diskette when the data was recorded (see Section
13.6). This error is usually caused by switching the disk drive on or off with a diskette
in it. It may also be part of a disk protection scheme.
24: READ ERROR
Byte decoding error. The disk controller has read an encoded byte from diskette for
which there is no hex equivalent. The 1541 drive does not check for this type of error.
25: WRITE ERROR
The disk controller found the desired header block and has written out the new data
block. However, when the data block was read back from the diskette, it did not
match what was just recorded. It is usually caused by a flaw on the diskette.
26: WRITE PROTECT ON
The disk controller has received a command to record some new information on the
diskette. However, the write protect notch in the diskette's jacket is covered by a
write protect tab. The information has not been written. The diskette has not been
affected in any way.
27: READ ERROR
The disk controller has moved the read/write head to the desired track and read the
header block of the first sector it has found to confirm that it was on the correct track
(this is called a SEEK operation). The header block information has been decoded.
348
When the checksum of the information in the header block was calculated, it did not
compute. This is usually caused by switching the disk drive on or off with a diskette in
it. It is occasionally used as part of a disk protection scheme.
28: WRITE ERROR
Long data block. The disk controller found the desired header block and has written
out the new data block. However, the SYNC mark at the start of the next sector's
header was not detected. The 1541 drive does not check for this type of error.
29: DISK ID MISMATCH
The disk controller has moved the read/write head to the desired track and read the
header block of the first sector it found to be sure that it was on the correct track (a
SEEK operation in preparation for a READ). The header block information has been
decoded. The header block checksum worked out properly. However, the two ID
characters recorded in the header when the diskette was formatted do not match the
master diskette ID stored in the disk drive's RAM memory. This is usually caused by
head alignment problems or changing diskettes and failing to initialize the new
diskette. It is frequently used as part of a disk protection scheme.
30: SYNTAX ERROR
General syntax error. The disk operating system cannot interpret the command sent
over the command channel. This error is usually caused by using too many file names
or using wild card characters improperly.
31: SYNTAX ERROR
Invalid command. The disk operating system does not recognize the command sent
over the command channel. For example, the DUPLICATE command, used on the
Commodore dual disk drive to make a backup copy of an entire diskette, is an invalid
command on the 1541.
32: SYNTAX ERROR
Long line. The disk command sent over the command channel is more than 58
characters long.
33: SYNTAX ERROR
Invalid file name. Pattern matching is used illegally in an OPEN or SAVE command.
34: SYNTAX ERROR
Missing file name. One or more file names were omitted from a disk command that
requires them. This error is usually the result of leaving a colon (:) out of a disk
command.
39: RLE NOT FOUND
An unusual error. It occurs only when using the 1541 utility loader (an & disk
command). The utility program named in the command was not listed in the diskette's
directory.
349
50: RECORD MOT PRESENT
An attempt has been made to use the POSITION disk command to access a record
beyond the end of a relative file. This error may be ignored if you are writing new
records into a relative file. The only time to be concerned about this error is when you
are attempting to read a record that should exist. Note that as soon as you POSITION
to a nonexistent record, the disk operating system automatically extends the length of
the relative file to include the record requested.
51: OVERFLOW IN RECORD
You just attempted to store too much data in a relative file record using a PRINT#
statement. The first part of the data has been written into the record but the extra
characters have been deleted to protect the next record in the file.
52: FILE TOO LARGE
The record number you specified in the POSITION disk command is so large that the
file will not fit onto the diskette.
60: NX/RITE FILE OPEN
An attempt has been made to read a file that has not been closed properly. This error
usually occurs when you attempt to read a file that has an asterisk (*) beside its name
in the directory (e.g., *SEQ, *PRG or *USR). To read a file of this kind you must
specify the modify (M) mode in the OPEN statement.
61: FILE NOT OPEN
An attempt has been made to access a file that has not been opened. This is a very
rare message. Normally the DOS simply ignores the request.
62: FILE NOT FOUND
There is no file name in the directory that matches the one specified in the OPEN,
LOAD, COPY or RENAME command. This error does not occur if the file name as
specified in a SCRATCH command does not match any name in the directory. In this
case, the DOS simply sends the 01, FILES SCRATCHED, message indicating that
zero files have been scratched.
63: FILE EXISTS
The file name to be added to the directory as the result of an OPEN, SAVE, COPY or
RENAME command matches an existing file name in the directory.
64: FILE TYPE MISMATCH
The file type (SEQ, PRG, USR or REL) specified in a disk command does not match
the file type in the directory entry for the file named in the command.
65: NO BLOCK
This message occurs only in response to aB-A (block allocate) disk command when
the block (sector) to be allocated is already allocated in the BAM. The track and sector
numbers returned along with this error message are supposed to be those of the next
350
available block (sector). Unfortunately, this feature of the B-A command does not
work properly on the 1541. If you get this error, VALIDATE the diskette because an
entire trackful of sectors will have been allocated incorrectly in the BAM. Luckily, no
sectors are freed as a result of this problem, so even if you do not notice it immediately,
your files are safe. However, you will run out of storage space very rapidly.
66: ILLEGAL TRACK AND SECTOR
An attempt was made to access a block (sector) whose track or sector number was
outside the allowed range. You may encounter this error (track 73, sector 1) when
reading an unclosed file using modify (M) mode.
67: ILLEGAL SYSTEM T OR S
A very rare error indicating that the system has attempted to access an illegal track
or sector.
70: NO CHANNEL
The requested channel is not available. A file you specified in an OPEN statement
was not opened. The most common reasons for a file not being opened are an incor-
rectly spelled file name (in read mode), a file type mismatch, a duplicate file name (in
write mode), or too many files already open. Only a few disk files can be opened
simultaneously. You cannot have five sequential files opened simultaneously as
specified on page 45 of the 1541 User's Manual. The correct, maximum, allowable
limits are:
Number of Number of Number of direct
SEQ files REL files access files TOTAL
2SEQ OREL OorlD-A 2or3
1SEQ 1REL OorlD-A 2or3
OSEQ 1REL 0to2D-A lto3
71: DIRECTORY ERROR
A very rare error. The BAM is not internally consistent. To cure the problem re-
initialize the diskette. Note that initializing the diskette will close all the data channels
that are open in the disk drive. The command channel is not affected.
72: DISK FULL
Either all the storage space on the diskette has been used, or the diskette directory is
full. The directory can only accomodate 144 file names. Note that the 1541 will close
the file properly if a disk full error is encountered during a save or replace operation.
However, the file will be incomplete/The only way to tell that this has happened is by
reading the error status. If you look at the directory, everything will look OK. A
program file will even verify as OK. Bui it is not all there! Beware! If a disk full error
is encountered while writing to or appending to a file, the file will not be closed
automatically. If you do not close it, you will have an unclosed file!
351
73.-CBMDOSV2.61541
When the 1541 disk drive is first turned on, this message is generated. To see it you
must check the error status as soon as you turn ON your drive.
74: DRIVE NOT READY
An unusual error that occurs mainly when using direct-access files or after doing a
drive reset ("U;"). If you encounter this error while using direct-access files, change
your OPEN statement to specify the buffer number explicitly. I had this problem
with the MOD ENTRY program and it would not go away until I changed OPEN
2,8,5,"#"toOPEN2,8,5,"#0"!
352
APPENDIX
D
SUMMARY OF DISK COMMANDS
Disk commands are usually sent to the 1541 using a PRINT# statement to print the
command to the command channel (previously opened as file# 15 in all our examples).
However, you can also combine the PRINT# statement with the OPEN statement or
use DOS 5.1 as indicated below.
SYNTAX:
PR I NT# 1 5 , "command " Usual form
OPEN 15, 8, 15, "command " Combined form
>command DOS 5.1 form
EXAMPLES:
PRINT#15, "NO: TEST DISKETTE, Tl "
PRINT#15, "V"
OPEN 15,8,15, "NO: TEST DISKETTE, Tl"
OPEN 15,8, 15, "V"
>no:test DISKETTE, Tl
NOTES:
1. Punctuation is important in these commands. Check it carefully.
2. The drive number that follows the command (e.g., NO) is always optional on the
1541 drive. It is only required when using a dual drive such as the Commodore
4040 drive.
3. Any housekeeping command may be abbreviated by using its first letter in place
of the full name of the command (e.g., N=NEW).
Housekeeping Commands
1. Full NEW of a diskette.
•
Prepares a new, blank diskette for first use.
May be used on a previously formatted diskette to erase errors.
Disk error aborts command, indicates flawed diskette.
Recovery is possible only if command aborted, see Chapter 11.
353
syntax:
NEWO: di skname, i d
examples:
print#15, "newo: test diskette, tl "
print#15, "no: test diskette, tl "
open 15,8, 15, "n: test diskette, t1":cl0se15
2. Short NEW of a diskette.
• Prepares a previously formatted diskette for reuse.
• The diskette ID is unchanged.
• Disk error aborts command, trouble on 18/0 or 18/1.
• Recovery is possible, see Chapter 11.
SYNTAX:
NEWO: di skname (Note: No diskette ID specified.)
EXAMPLES:
PR I NT# 15," NEWO : TEST D I SKETTE "
PRINT#15,"N0:TEST DISKETTE"
OPEN 15,8, 15, "N: TEST DISKETTE": CLOSE 15
3. VALIDATE a diskette.
• Creates a new BAM based on the files actually stored.
• Disk error aborts command, BAM is unchanged.
SYNTAX:
VALIDATEO
EXAMPLES:
PRINT#15, "VALIDATEO"
PRINT#15, "VO"
OPEN 15,8, 15, "V":CL0SE15
4. SCRATCH a file.
WARNING:
Do not scratch an unclosed file! (*SEQ, *PRG, *USR).
• Eliminates unwanted file from the directory.
• Wild cards permitted to scratch several files.
• Disk error aborts command, BAM may be incorrect.
• Recovery is possible, see Chapter 11.
SYNTAX:
SCRATCHO: -f i 1 ename
354
examples:
pr i nt# 1 5 , " scr atcho : old file"
PRINT#15,"S0:0LD FILE"
OPEN 15,8, 15, "S: OLD FILE":CL0SE15
5. COPY a file.
• Makes a second copy of a file on the same diskette.
• No wild cards permitted.
• Disk error aborts command, BAM may be incorrect.
• Does not work with relative files.
SYNTAX :
COPYO: backupfile=0: old-file
EXAMPLES:
PRINT#15, "COPYO: 6AME/BU=0: SAME"
PRINT#15, "co:game/bu=o:gamem
OPEN 1 5 , 8 , 1 5 , " C : GAME/BU=SAME " : CLOSE 1 5
6. CONCATENATE two to four files.
• Makes new, combined file composed of contents of first file followed by contents
of second, followed by contents of third...
• No wild cards permitted.
• Does not merge two BASIC programs.
• Disk error aborts command, BAM may be incorrect.
• Does not work with relative files.
SYNTAX :
COPYO: combofile=0:fi lei, 0:*ile2,0:file3
EXAMPLES:
PR I NT# 15," COPYO : OURS=0 : YOURS , O : M I NE "
PRINT#15, "CO: OURS=0: YOURS, O: MINE"
OPEN 15, 8, 15, "C: OURS= YOURS, MINE" : CL0SE15
7. RENAME a file.
• Changes the file name in the directory entry.
• No wild cards permitted.
• Disk error aborts command, trouble on track 18.
• Directory may be fixed, see Chapter 11.
SYNTAX:
RENAMED : new* i 1 e=ol df i 1 e
EXAMPLES:
PRINT#15,"RENAME0:NEW NAME=OLD NAME"
355
PRINT#15, "RO:NEW NAME=OLD NAME"
OPEN 15,8,15:PRINT#15, "R:NEW NAME=OLD NAME-
CLOSE 15
8. INITIALIZE a diskette.
• Read track 18, sector 0 into disk RAM ($0700-FF).
• Used to prevent 29, DISK ID MISMATCH, errors.
• Disk error aborts command, trouble on track 18, sector 0.
• Recovery is possible, see Chapter 11.
SYNTAX:
INITIALIZEO
EXAMPLES:
PRINT#15, "INITIALIZEO"
PRINT#15,"I0"
OPEN 15,8, 15, "I":CL0SE15
Direct-Access Commands
These commands are used mainly by experienced programmers to create specialized
utility programs. They are not needed for normal file handling.
1. BUFFER-POINTER (B-P)
• Position to the specified byte in the data buffer to prepare to read or write data
to the buffer.
• A direct-access channel must have been opened previously.
SYNTAX:
"B-P: "channel #;byte#
EXAMPLES:
OPEN 1,8,5, "■#"
PRINT* 15, "B-P: 5 O" Position to byte zero of buffer.
OPEN 3,8,4, "#1"
PR I NT# 15," B-P : 5 25 " Position to byte 25 of data buffer #1.
2. BLOCK-READ (B-R)
WARNING:
The B-R command does not work properly. Use Ul below.
BLOCK-READ (Ul:)
• Read specified track and sector into the 1541's RAM memory.
• A direct-access channel must have been opened previously.
356
syntax:
"Ul : "channel #; dri ve#; track#; sector*
EXAMPLES:
OPEN 1,8,5,"*"
PRINT#15, "Ul:5 0 18 1"
Read track 18, sector 1, from drive zero into the buffer assigned to the file using
data channel five.
OPEN 3,8,4, "#1"
TR=15:SE=10
PRINT#15,"Ul:4 0"TR?SE
PRINT#15, "B-P:4 32"
6ET#3, A*
Read track 15, sector 10 from drive zero into data buffer (one) assigned to the file
using data channel four. Then position to byte 32 and read it.
3. BLOCK-WRITE (B-W)
WARNING:
This command does not work properly. Use U2 below.
BLOCK-WRITE (U2:)
• Write contents of data buffer to specified track and sector.
• A direct-access channel must have been opened previously.
SYNTAX:
"U2: "channel #; dr i ve#; track*; sector*
EXAMPLE:
OPEN 1,8,5,"*"
PR I NT* 1 5 , " B-P : 5 O " Position to start of buffer.
PR I NT* 1 , M* ; Store data in buffer.
PRINT*15, "Ul:5 0 18 1"
Record contents of the buffer assigned to the file using data channel five on drive
zero, track 18, sector 1.
4. BLOCK-ALLOCATE (B-A)
Mark the specified track and sector as allocated (in use) in the diskette's BAM.
WARNING:
This command works correctly only if the designated block is free in the BAM. If
the block is already allocated, an entire sector full of blocks will be allocated
and the BAM will be incorrect.
357
syntax:
"B-A: "drive#;track#?sector#
EXAMPLES:
PRINT#15,"B-A:0 17 4" Allocate track 17, sector 4.
TR=12:SE=5
PR I NT# 15," B-A : O " TR ; SE Allocate track 12, sector 5.
5. BLOCK-FREE (B-F)
Mark the specified track and sector as free (available for use) in the diskette's
BAM.
SYNTAX:
"B-F: "drive#;track#5sector#
EXAMPLES:
PRINT#15, "B-F:0 17 4" Free track 17, sector 4.
TR=12:SE=5
PR I NT# 15," B-A : 0 " TR ; SE Free track 12, sector 5.
6. BLOCK-EXECUTE (B-E)
Read the contents of the track and sector specified into a data buffer disk RAM
and then begin executing the machine language routine that begins at the start of
this buffer. The 1541's microprocessor will be in file manager mode as it executes
the code.
syntax:
"B-E: "channel #;drive#;track#; sector*
EXAMPLE:
OPEN 1,8,5, "#1"
PRINT#15, "B-E: 5 O 24 3"
Read the contents of track 24, sector 3 into data buffer #1 (see OPEN) and begin
executing the code that starts at memory location $0400 in RAM memory (data
buffer #1 starts at $0400).
7. MEMORY-READ
• Prepares to read the contents of the 1541's RAM or ROM memory.
• Contents are actually read using GET# to read the command channel.
NOTE:
The syntax given in the 1541 User's Manual is incorrect.
358
syntax:
MM-R"CHR*(adr lo);CHR*(adr hi)
HM-R"CHR*(adr lo);CHR*(adr hi); CHR* (how many)
EXAMPLES:
PRINTttl'S, MM-R"CHR*(2) ;CHR*(7)
GET#15,A*
Read one byte of memory at $0702.
PRINT#15,"M-R"CHR*(0);CHR*(240);CHR*(32)
FOR K=0 TO 31
GET#15,MEM*(K)
NEXT
Read 32 bytes of memory starting at $0702.
8. MEMORY-WRITE
• Writes bytes in M$ directly into the 1541's RAM memory.
• The maximum number of bytes per "M-W" command is 32.
NOTE:
The syntax given in the 1541 User's Manual is incorrect.
SYNTAX:
"M-W"CHR*<adr lo);CHR*(adr hi); CHR* (how many),M*
EXAMPLES:
PRINT#15, "M-W"CHR* <3) | CHR* (O) 5 CHR* ( 1 ) ; CHR* (176)
Stores CHR$(176) into memory at $0003.
FOR K=0 TO 31
READ X:C*=C*+CHR*(X)
NEXT
PRINT#15, "M-W"CHR* (O) ; CHR* (3) 5 CHR* (32) 1 M*
Stores 32 bytes in memory starting at $0300.
9. MEMORY-EXECUTE
Forces the 1541's microprocessor (in file manager mode) to start executing the
machine language routine that begins at the location specified in the disk memory.
NOTE:
The syntax given in the 1541 User's Manual is incorrect.
359
syntax:
"M-E"CHR*(adr lo);CHR*(adr hi)
EXAMPLES:
PRINT#15, "M-E"CHR* (O) » CHR* <5>
Execute machine language routine that begins at memory location $0500 in RAM
memory.
10. RESET DRIVE ("U;", "UI", or "U9")
• Reset the 1541 drive to its power-on state.
• Note that the drive motor will not start up.
NOTE:
The "U" command given in the 1541 User's Manual is incorrect.
SYNTAX:
"U? " or "UI" or "U9"
EXAMPLES:
PRINT#15, "U; "
PRINT#15,"UI"
PRINT#15, "U9"
360
APPENDIX
E
DISK UTILITY PROGRAMS
The disk utility programs listed below are general purpose programs. More
specialized utility programs are listed in the appropriate section of the book. The
Table of Contents contains a complete list of the application and utility programs
included herein.
WARNING
The programs listed below are very powerful disk utilities. They have been designed
for ease of use and relatively foolproof operation. However, any tool may be misused.
Be sure to practice using a utility on a test diskette before you use it with one of your
important or valuable diskettes.
Summary of Programs Listed in Appendix E
Program Name Function
FULL DIRECTORY Expanded directory listing.
CONFIRM ALL FILE Checks to ensure all disk files are intact.
BACKUP 1541 Backup a diskette. Does not abort on errors.
MAKE FAST COPY BASIC program to make a working copy of the
FAST COPY program from Chapter 10.
EDIT T&S Hex or ASCII editing of any disk block.
MOD DISK NAME Modify any diskette's name.
FIX BAD DIR Recover files after a short NEW or if the directory
has been damaged.
THE UNSCRATCHER Recover any type of scratched files.
PRG HEX DUMP Analysis of BASIC program showing links, etc.
PRG ANALYSIS Detailed analysis of BASIC program file.
***********************
1. PROGRAM: FULL DIRECTORY
***********************
Produces an expanded directory listing for a diskette showing file length, file
type, file name, track and sector where first block is stored, and the load address
of the file. See Section 11.1 for instructions.
Compatible with C-64 or VIC-20 and 1541.
361
0 REM 1541 USER'S GUIDE APPENDIX E
1 REM COPYRIGHT: G. NEUFELD, 1964
2 :
3 REM DISPLAY EXTENDED DIRECTORY
4 :
lOO PRINT" {CLR> 1541 FULL DIRECTORY": A*="
S"
HO DIM FX*(7),DB(1>
120 INPUT" <DOWN> OUTPUT TO SCREEN OR PR IN
TER <S/P>";A*
130 DN=3: IF A*="P" THEN DN=4
140 OPEN 3,DN
150 OPEN15,8,15, "10"
160 OPEN 2,8,6, "#"
170 H*="0123456789ABCDEF"
180 Z*=CHR*<0>
190 FOR K=0T05:READFX*(K):NEXT
200 DATA DEL, SEQ, PRG,REL,USR, *w>
210 OPENl,8,5, "*6,P,R"
220 GET#1 , A*: IFA*=" "THENA*=Z*
230 DOS=ASC(A*>
240 F0RK=4T0144: GET#1 , A*: NEXT
250 FORK= 1 45T0 1 60 : GET# 1 , A* : I F A*= " " THENA«
=Z*
260 N*=N*+A*:NEXT
270 FORK= 1 6 1 TO 1 62 : GET* 1 , A* : NEXT
280 GET#1,A*: IFA*=""THENA*=Z*
290 1 1 =ASC ( A* > : 1 1 *=A*
300 GET# 1 , A* : I FA*= " " THENA*= Z *
310 I2=ASC (A*) : I2*=A*
320 PRINT#3, "<CLR>DISK NAME: "N*
330 PRINT#3, "DOS TYPE: "DOS
340 PRINT#3, "COSMETIC DISK ID: ";
350 IF Il>31 AND IK96 THEN PRINT#3, II*;
: GOT0370
360 PRINT#3, " ";
370 IF I2>31 AND I2<96 THEN PRINT#3, 12*;
: GOT0390
380 PRINT#3, " ";
390 PRINT#3, " ASCII ="I1;I2
400 PRINT#3,"CD0WN>BLKS FILE NAME
TYPE LINK START"
410 PRINT#3, "
420 FOR K=165T0256:GET#1,A*:NEXT
430 GET# 1 , A* : I FA*= " " THENA*=Z*
440 IF STOO GOTO 780
450 FT=ASC<A*>
460 FT*=FX*(7ANDFT>
470 IF FT<128 THEN FT*=" CRVS> "+FT*+" CROF
F3"
480 GET#1 , A*: IFA*=" "THENA*=Z*
490 TR=ASC(A*>
500 TR*=RIGHT*<" "+STR*<ASC (A*) ) ,4)
510 GET#1,A*:IFA*=""THENA*=Z*
520 SE=ASC<A*>
530 SE*=RIGHT*(" "+STR*<ASC<A*> > ,3) +"
":N*=""
540 IF TR=0 AND SE=0 GOTO 780
550 PRINT#15, "Ul:6 0"5TR;SE
560 PRINT#15?"B-P:6 2"
570 GET#2 , A* : I F A*= " " THEN A*= Z *
5S0 A=ASC(A*> :H=INT<A/16> :L=A-16*H
590 HX*=MID* <H*„ H+l , 1 ) +MID* (H*, L+l , 1 )
600 GET#2,A*:IFA*=""THENA*=Z*
610 A=ASC<A*> :H=INT<A/16> :L=A-16*H
620 HX*="*"+MID* <H$, H+l , 1 ) +MID* (H*, L+l , 1
>+HX*
630 FOR K=1T016:GET#1,A*:IFA*=,,,,THEN A*=
Z*
640 N*=N*+A*:NEXT
650 N*=" "+N*+"
660 FORK=lT09:GET#l,A*:NEXT
670 GET# 1 , A* : I FA*= " " THENA*=Z*
680 BL=ASC(A*>
690 GET#1 , A*: IFA*=" "THENA*=Z*
700 BL=BL+256*ASC<A*>
710 IF FT<128 THEN DB<0>=DB(0>+BL
720 IF FTM27 THEN DB( 1 >=DB ( 1 ) +BL
730 BL*=R I GHT* < " " +STR* < BL> , 3 ) + " "
740 PRINT#3, BL*5 N*5 FT$; TR*5 SE$? HX*
750 ZZ=ZZ+l:IFZZ/8<>INT<ZZ/8) THEN GET#1
,A*:6ET#1,A*
760 GOTO 430
770 :
780 PR I NT# 1 5 ., " M-R " CHR* < 250 > CHR* < 2 > CHR* ( 3
>
790 6ET#15,A*:BF=ASC(A*+Z*>
800 GET#15,A*
810 GET#15, A*: BF=BF+256*ASC ( A*+Z*>
820 PRINT" tDOWN> BLOCKS FREE PER DIRECTOR
Y ="664-DB(l)
830 PR I NT "BLOCKS FREE ACCORDING TO BAM="
BF
840 PR I NT "SCRATCHED BLOCKS="DB<0)
850 IF BF >( 664-DB ( 1 ) > THENPR I NT " < DOWN} CRO
SSED FILES LIKELY"
860 IF BF<(664-DB(1))THENPRINT"£D0WNJ"66
4-DB(l)-BF" NON-FILE BLOCKS ALLOCATED"
870 CLOSE 1 : CLOSE2 : CLOSE3 : CLOSE 1 5 : END
*************************
2. PROGRAM: CONFIRM ALL FILE
*************************
Checks all files, BAM and directory for errors. The location of any collisions
between files are displayed. Output may be sent to the screen or printer. Optional
display of file chains of all files and a SUPER BAM showing where all files are
stored and highlighting any BAM errors. See Section 11.1 for complete
instructions.
Compatible with C-64, VIC-20, 4032 and 8032 computers, 1541, 2031 and 4040
disk drives (use drive zero).
0 REM 1541 USER'S GUIDE APPENDIX E
1 REM COPYRIGHT: G. NEUFELD, 1984
2 :
3 REM CHECK ALL FILES FOR ERRORS
4 :
100 PRINT" CCLRJCONFIRM A DISKETTE": PRINT
110 DIM M-/.(35,20),N*(144),S*(35),BM7.(35,
20)
1 20 B*= " " : Z *=CHR* ( O ) : SE*= " ROTCES "
130 FOR K=0 TO 7: READ FT*(K) :FT* <K>=". . .
."+FT*(K):NEXT
140 DATA DEL, SEQ,PRG,USR,REL, ???,???,???
150 FOR K=0 TO 35
160 T=INT(K/10) :S*(K)=CHR*(48+T)+CHR*(48
+K-10*T)
170 NEXT
180 INPUT "OUTPUT TO SCREEN OR PRINTER S
CLEFT 3>";P*
190 DN=3:IF(ASC<P*)AND127)=80 THEN DN=4
200 OPEN 6,DN
210 PRINT: INPUT "DISPLAY FILE CHAIN <Y/N
) Y CLEFT 3>";P*
220 FC=1 : IF (ASCtP*) AND 127)089 THEN FC=0
230 PRINT: PRINT" INSERT DISK IN DRIVE"
240 PRINT"PRESS CRVS> RETURN <ROFF> WHEN R
EADY"
250 GET A*: IF A*OCHR*(13) GOTO 250
260 PR I NT: PR I NT "READING BAM <30 SECOND P
AUSE) "
270 OPEN 15,8,15, "10"
280 GOSUB 1320
290 OPEN 1,8,4, "#"
300 GOSUB 1320
310 OPEN 2,8,5, "#"
320 GOSUB 1320
364
330 GOSUB 1430
340 PRINT#15, "Ul:4 O 18 1 "
350 dp=o:nf=o
360 :
370 REM MAIN LOOP
380 GOSUB SOOZREM TRACE NEXT ENTRY
390 IF FX=1 GOTO 380
400 :
410 CLOSE l:CL0SE2
420 INPUT#15,E,E*,T,S
430 CLOSE 15
440 PRINT#6, :PRINT#6, "ALL FILES HAVE BEE
N CHECKED"
450 input" <:down> print SUPER BAM <Y/N> Y
{LEFT 3>";X*
460 IF ASC<X*>=89 THEN GOSUB 1720
470 CLOSES: END
480 :
490 REM: FIND NEXT FILE IN DIRECTORY
500 FD= 1 : SF=0 : PR I NT# 15," B-P : 4 " 2+32*DP
510 GET#1,A*:FT=135ANDASC<A*+Z*>
520 IF FT > 127 GOTO 540
530 SF=1
540 NF=NF+1
550 GET# 1 , A* : TR=ASC ( A*+Z* >
560 GET* 1 , A* : SE=ASC < A*+Z* )
570 FOR K=1T016
580 GET#1,A*:IF A«=B« THEN A*=Z*
590 N*(NF)=N*(NF)+A*
600 NEXT
610 IF SF=0 GOTO 660
620 IF TR=0 AND SE=0 THEN NF=NF-l:FD=OlG
0T0900
630 PRINT: PRINT" CRVS> SCRATCHED FILECROFF
> ";N*(NF>
640 INPUT "TRACE IT (Y/N) NO-EFT 3>";R*
650 IF R*<>"Y" THEN FD=0:G0T0900
660 PRINT#6,:PRINT#6, "TRACING: "N*<NF)FT
*(7ANDFT)
670 :
680 REM TRACE MAIN FILE
690 GOSUB 1090
700 IF FT0132 GOTO 780
710 :
720 REM TRACE SIDE SECTOR FILE
730 PRINT: PRINT"SIDE SECTOR FILE: ";
740 GET# 1 , A* : TR=ASC ( A*+Z* )
750 6ET# 1 , A* : SE=ASC < A*+Z* )
760 GOSUB 1100
770 :
365
780 FD= 1 : PR I NT# 1 5 , " B-P : 4 " 30+32*DP
790 GET#1,A*:DB=ASC(A*+Z*>
800 GET# 1 , A* : DB=DB+256*ASC < A*+Z* )
81 O IF FC=0 GOTO 850
820 IF DN=3 AND (ZZ/6 <> INT<ZZ/6>> THEN
PRINT#6,
830 IF DN=4 AND (ZZ/13 <> INT(ZZ/13>) TH
EN PRINT#6,
840 IF ZZODB THEN PRINT#6, " <RVS>";
850 PRINT#6, ZZ" BLOCKS FOUND. "DB" IN DIREC
TORY"
860 DP=DP+i
870 IF DP<8 GOTO 1030
880 :
890 REM: GET LINK TO NEXT DIR SECTOR
900 DP=0
910 PRINT#15. "B-P: 4 O"
920 GET# 1 , A* : TD=ASC < A*+Z* )
930 GET#1,A*:SD=ASC(A*+Z*)
940 :
950 REM: IF NO LINK, DONE!
960 IF TD=0 THEN FX=0: RETURN
970 :
980 REM: READ NEW DIR SECTOR
990 PRINT#15, "Ul:4 0"TD;SD
lOOO GOSUB 1320
1010 :
1020 REM: IF NO FILE FOUND, TRY AGAIN
1030 IF FD=0 GOTO 500
1040 :
1050 FX=l: RETURN
1060 :
1070 REM: TRACE FILE BY LINKS
1080 :
1090 ZZ=0
1100 IF M7.(TR,SE)O0 THEN GOSUB 1250:RET
URN
1110 M7.(TR,SE)=NF
1120 ZZ=ZZ+l:IF FC=0 GOTO 1170
1130 IF BM%(TR,SE)=1 THEN PRINT#6, " <RVS>
SI m
%
1140 PRINT#6.S*<TR) ", "S*(SE) " £ROFF> ";
1150 IF DN=3 AND <ZZ/6 = INT(ZZ/6>) THEN
PRINT#6,
1160 IF DN=4 AND (ZZ/13 = INT<ZZ/13>) TH
EN PRINT#6,
1170 PRINT#15, "Ul:5 0"TR;SE
1180 GOSUB 1320: IF E>19 THEN RETURN
1190 PRINT#15, "B-P: 5 O"
1200 GET#2,A*:TR=ASC(A*+Z*>
366
1210 GET#2,A*:SE=ASC(A*+Z*>
1220 IF TR>0 AND TR<35 GOTO 1100
1230 RETURN
1240 :
1250 PRINT#6:PRINT#6, "{RVS>COLLISION WIT
H ,'N*(My.(TR.SE>)
1260 PRINT#6, "COLLISION ON TRACK "TR" SECT
OR"SE
1270 RETURN
1280 :
1 290 CLOSE 1 : CL0SE2 : CL0SE6
1300 INPUT#15,E,E*,T,S:CL0SE15:END
1310 :
1320 INPUT#15,E,E*,T,S
1330 IF E<20 THEN RETURN
1340 PRINT: PRINT-DISK ERROR: "e;e*;t;s
1350 IF EF=1 THEN EF=0: RETURN
1360 PR I NT: PR I NT "PRESS C TO CONTINUE OR
Q TO QUIT"
1370 GET A*: IF A*="" GOTO 1370
1380 IF A*="C" THEN RETURN
1390 IF A*="Q" GOTO 1290
1400 GOTO 1370
1410 :
1420 REM: READ THE BAM
1430 PRINT#15, "Ul:4 O 18 O"
1440 EF=l:GOSUB 1320: IF E>19 GOTO 1290
1450 PRINT#15, "B-P:4 144"
1460 F0RK=1T024:GET#1,A*:IFA*=""THEN A*=
CHR*(0)
1470 DN*=DN*+A*:NEXT
1480 PRINT#6, "<CLR> CONFIRMING {RVS> "DN*
1490 PRINT#6: PR I NT#6, "CHECKING DISKETTE
BAM"
1500 PRINT#15,"B-P:4 4"
1510 FOR TR=1TD35
1 520 GET# 1 , A* : T7.=ASC ( A*+Z* ) : B7.=0
1530 NS=20+2* (TR>17) + (TR>24) + (TR>30)
1540 GET#1,A*:A=ASC(A*+Z*)
1550 FOR S=0T07:IFAAND2~STHENBM7.<TR,S>=1
:B%=B7.+1
1560 NEXT S
1570 get#i„a*:a=asc<a*+z*>
1580 for s=ot07:ifaand2asthenbm%<tr!>s+8>
=l:B7.=B7.+l
1590 NEXT S
1 600 GET# 1 , A* : A=ASC ( A*+Z* )
1610 FOR S=OTO < NS- 1 6 ) : I F AAND2 ASTHENBM7. < T
R, S+16) =1 : B7.=B7.+1
367
1620 NEXT S
1630 IF B%=T% GOTO 1660
1640 PRINT#6?" INTERNAL €SHIFT-B> CSHIFT-
amshift-mj error, track -tr:be=i
1650 print#6, " master blocks free="t7."
actual count="b%
1660 NEXT TR
IF BE=0 THEN PRINT#6, "BAM IS INTERN
CONSISTENT"
RETURN
1670
ALLY
1680
1690
1700
1710
1720
M OF
1730
#6
REM DISPLAY SUPER BAM
IF DN=3 THEN
{RVS> "DN*
IF DN=4 THEN PRINT#6,CHR*(19>
CCLRJ SUPER BAM OF €RVS> "DN*
PR I NT#6 , " CCLR> SUPER BA
PRINT
FOR SE=20T00 STEP-1
IF SE>5 THEN SX*=" ":G0T01770
SX*=MID* (SE*, SE+1 , 1 )
PRINT#6, SX*; RIGHT* (STR* (SE) , 2) " CCOM
IF SE=20 AND TR>17 THEN TR=35:G0T0
IF SE=19 AND TR>17 THEN TR=35:G0T0
IF SE=18 AND TR>24 THEN TR=35:GOTO
IF SE=17 AND TR>30 THEN TR=35:G0T0
1740
1750
1760
1770
-Q> " ;
1780 FOR TR=1T035
1790
1880
1800
1880
1810
1880
1820
1880
1830 IF TR=18 AND BM7.(TR,SE>=0 THEN PRIN
T#6 ,"©";: GOTO 1 880
1840 IF M7.(TR,SE>=0 AND BM5i(TR,SE>=0 THE
N PRINT#6, " CRVS> <COM-+> CROFFJ " ; : GOTO 1880
1850 IF M7.(TR.,SE)=0 AND BM7.(TR, SE>=1 THE
N PRINT#6, " CCQM-+} " ; : G0T01880
1860 IF BM7.(TR?SE)=1 THEN PRINT#6, "{RVSJ
II a
1870 PRINT#6, CHR* (64+ <31ANDM7. (TR, SE) ) ) " €
ROFF> " ;
1880 NEXT TR:PRINT#6:NEXT SE
1890 PRINT#6, " tCOM-ZJCCOM-E 35}"
1900 PRINT#6," 1234567890123456789012
3456789012345"
1910 IF DN=4 GOTO I960
1920 PRINT#6, " TRACK PRESS CRVS>
RETURNCROFF* TO END";
1930 GET A*: IF A*<>" "GOTO 1930
1940 GET A*: IF A*OCHR*(13)G0T01940
1950 RETURN
1960 PRINT#6, " TRACK"
1970 PRINT#6:PRINT#6, "KEY TO SUPER BAM":
PRINT#6
1980 PRINT#6, "@ = DIRECTORY (TRACK 18 ON
LY>"
1990 FOR K=l TO NF
2000 PRINT#6,CHR*(64+(31ANDK))" = ";N*(K
>
2010 NEXT
2020 RETURN
********************
3. PROGRAM: BACKUP 1541
********************
Makes a duplicate copy of a diskette. It takes about 25 minutes to backup a
diskette. Six full tracks are read into the C-64's RAM from the master disk and
then written out to the clone. A chime sounds to indicate operator attention is
required. The main advantage of this backup program is that it will not abort on
read or write errors. A summary of the sectors not copied is provided.
Compatible with C-64 computer and 1541 disk drive only.
0 REM 1541 USER'S GUIDE APPENDIX E
1 REM COPYRIGHT: G. NEUFELD, 1984
2 :
3 REM BACKUP FOR THE 1541 SINGLE DISK D
RIVE
4 REM WILL NOT ABORT ON READ OR WRITE E
RRORS
5 :
100 REM C-64/ 1541 BACKUP
110 PRINT" CCLRJC-64/ 1541 BACKUP"
120 P=PEEK (8*4096) +l: IF P>255 THEN P=0
130 POKE 8*4096,P
140 IF PEEK (8*4096) OP THEN PR I NT "NO CAR
TRIDGES ALLOWED": END
150 POKE 55,0: POKE 56, 32: POKE 251,0:CLR
160 SID=54272
170 FOR K=0 TO 28: POKE SID+K,0:NEXT
180 POKE SID+24, 15
190 POKE SID, 34: POKE SID+1,75
200 POKE SID+5, 16: POKE SID+6,249
210 DIM MT(50),MS(50),EN(50),EM*(50)
220 FOR K=0 TO 37: READ X:POKE 828+K,X:NE
XT
230 GOSUB 830
240 OPEN 15,8, 15, "10"
369
250 GOSUB 1010: IF ER=1 THEN STOP
260 OPEN 1,8.5, "#"
270 PRINT#15, "Ul:5 O 18 O"
280 GOSUB 1010ZIF ER=1 THEN STOP
290 REM DIG OUT DISK NAME
300 PRINT#15, "B-P:5 144"
310 DN*=""
320 FOR K=l TO 16
330 GET#1,A*:IF A*="" THEN A*=CHR*(0)
340 DN*=DN*+A$
350 NEXT
360 REM DIG OUT ID
370 PRINT#15, "B-P:5 162"
380 GET#1,A«:IF A*="" THEN A*=CHR*<0)
390 ID«=A«
400 GET#1,A*:IF A*="" THEN A*=CHR*(0)
410 ID*=ID*+A*
420 CLOSE 1
430 PRINT" DISK NAME = "DN*
440 PRINT" DISK ID = "ID*
450 INPUT" (DOWN? CONTINUE (Y/N) YXLEFT 3
> " ; R*
460 IF R*<>"Y" THEN CLOSE 15: END
470 PRINT" CUP 3J": GOSUB 860
480 PRINT" FORMATTING DISKETTE"
490 PRINT#15, "NO: "+DN*+", "+ID*
500 GOSUB 1010:IF ER=1 THEN STOP
510 :
520 GOSUB 830: PRINT" READING TRACKS 1-6
" : T l = i : T2=6: gosub l 070
530 GOSUB 860: PRINT" WRITING TRACKS 1-6
": GOSUB 1220
540 :
550 GOSUB 830: PRINT" READING TRACKS 7-1
2":T1=7:T2=12:G0SUB 1070
560 GOSUB 860: PRINT" WRITING TRACKS 7-1
2": GOSUB 1220
570 :
580 GOSUB 830: PRINT" READING TRACKS 13-
18" :T1=13:T2=18: GOSUB 1070
590 GOSUB 860: PRINT" WRITING TRACKS 13-
18": GOSUB 1220
600 :
610 GOSUB 830: PRINT" READING TRACKS 19-
24" :T1=19:T2=24: GOSUB 1070
620 GOSUB 860: PRINT" WRITING TRACKS 19-
24": GOSUB 1220
630 :
640 GOSUB 830: PRINT" READING TRACKS 25-
30 " : T 1 =25 : T2=30 : GOSUB 1 070
370
650 BOSUB 860: PRINT" WRITING TRACKS 25-
30":G0SUB 1220
660 :
670 GOSUB 830: PR I NT" READING TRACKS 31-
35 " : T 1 =3 1 : T2=35 : GOSUB 1 070
680 GOSUB 860: PRINT" WRITING TRACKS 31-
35": GOSUB 1220
690 :
700 PRINT" {DOWN* BACKUP COMPLETE"
710 POKE SID, 4: POKE SID+1,49
720 FOR L=l TO 10
730 POKE SID+4, 33: FORK=lT010: NEXT: POKE S
ID+4, 32: F0RK=1T01000: NEXT
740 NEXT
750 CLOSE 15
760 IF NE=0 THEN PRINT" CDOWN>ALL SECTORS
COPIED": END
770 PRINT" CDOWN* THERE WERE "NE" SECTORS NO
T COPIEDCDOWNJ"
780 FOR K=l TO NE
790 PRINT"T="MT <K) "S="MS <K> EN <K) EM* (K)
800 NEXT
810 END
820 :
830 PRINT" {DOWN> INSERT MASTER DISKETTES
OWN* "
840 POKE SID, 8: POKE SID+1,98
850 GOTO 890
860 PRINT" {DOWN} INSERT DESTINATION DISKE
TTECDOWNJ"
870 POKE SID, 16: POKE SID+1,195
880 GET A*: IF A*<>"" GOTO 880
890 PRINT" {UPJPRESS {RVS>RETURN{ROFF> WH
EN READY"
900 POKE SID+4, 33: FORK=lT010: NEXT: POKE S
ID+4, 32
910 FOR K=l TO 50
920 GET A*: IF A*=CHR*(13) GOTO 980
930 NEXT
940 PRINT" OJPJCRVS3 PRESS *ROFF> RETURN CRV
S3 WHEN READY"
950 FOR K=l TO 50
960 GET A*: IF A*=CHR*<13) GOTO 980
970 NEXT: GOTO 890
980 PRINT" {UP*
UP3- "
990 RETURN
1000 :
101O INPUT#15,EN,EM*,ET,ES
1020 IF EN<20 THEN ER=0: RETURN
371
1030 PRINT" {DOWN} DISK ERROR -
1040 printen;em*;et;es
1050 ER=l: RETURN
1060 :
1070 PK=32
1080 OPEN 1,8,5, "#"
1090 FOR T=T1 TO T2
1 100 NS=20+2* (T>17) + (T>24) + (T>30)
lllO FOR S=0 TO NS
1120 PRINT#15, "Ul:5 0"T;S
1130 GDSUB 1380
1140 PRINT#15, "B-P:5 O"
1150 POKE 252,PK:PK=PK+1
1160 SYS 828 :REM READ 256 BYTES
1170 NEXT S
1180 NEXT T
1190 CLOSE 1
1200 RETURN
1210 :
1220 PK=32
1230 OPEN 1,8,5, "#"
1240 FOR T=fl TO T2
1250 NS=20+2*(T>17)+<T>24)+(T>30>
1260 FOR S=0 TO NS
1270 PRINT#15, "B-P:5 O"
1280 POKE 252,PK:PK=PK+1
1290 SYS 847 :REM WRITE 256 BYTES
1300 PRINT#15, "U2:5 0"T;S
1310 GOSUB 1380
1320 A=FRE<0)
1330 NEXT S
1340 NEXT T
1350 CLOSE 1
1360 RETURN
1370 :
1380 INPUT#15,EN,EM*,ET,ES
1390 IF EN<20 THEN RETURN
1400 PRINT"TRACK,,ET"SECTOR"ES,,NOT COPIED
It
1410 PRINT" "EN;em*;et;es
1420 NE=NE+l:IF NE>50 THEN NE=50
1430 IF NE<50 THEN MT(NE)=ET:MS(NE)=ES:E
N ( NE ) =EN: EM* ( NE ) =EM*
1440 RETURN
1450 :
1460 DATA162, 1, 32,198,255,160, O
1470 DATA 32,207,255,145,251,200,208
1480 DATA248, 32,204,255, 96,162, 1
1490 DATA 32,201,255,160. 0,177,251
1500 DATA 32,210,255,200,208,248, 32
1510 DATA204,255, 96
372
PROGRAM: MAKE FAST COPY
A BASIC program that creates a working copy of the machine language FAST
COPY program given at the end of Chapter 10. To help eliminate typing errors,
the program checks each of the blocks of data before creating the disk file. The
FAST COPY program allows you to copy a file from one diskette to another
quickly and easily. Files up to about 120 blocks long may be copied. If you don't
waste time when swapping diskettes, copying takes about one second per block. It
will not copy relative files.
Compatible with C-64 computer and 1541 disk drive only.
0 REM 1541 USER'S GUIDE APPENDIX E
1 REM COPYRIGHT: G. NEUFELD, 1984
2 :
3 REM CREATE A WORKING COPY OF THE "FAS
T COPY" PROGRAM
4 :
lOO REM FAST COPY FOR THE C-64 AND 1541
110 REM MAKES PGM FILE OF FAST COPY PROG
RAM
120 DIM CKC20)
130 :
140 REM READ IN CHECKSUM VALUES
150 FOR K=l TO 20:READ CK(K):NEXT
160 :
170 REM CHECKSUM VALUES
180 DATA 3728,4805,4391,3477,4715
190 DATA 5747,4400,4743,5805,5288
200 DATA 4852,5841,3886,3485,2016
210 DATA 2024,2350,2477,2024, 195
220 :
230 REM CHECK FOR TYPING ERRORS
240 FOR BLK=1 TO 20:SUM=0:NV=35: IF BLK=2
0 THEN NV=4
250 FOR VL=1 TO NV:READ X:SUM=SUM+X: NEXT
260 IF SUMOCK(BLK) THEN FL=1:PRINT"ERR0
R IN BLOCK "BLK
270 NEXT
280 IF FL=1 THEN STOP
290 :
300 REM DUMP CHECKSUM VALUES
310 RESTORE: FOR K=1T020:READ x:next
320 :
330 PRINT" <DOWN> INSERT DISK FOR STORING
COPY PROGRAM"
340 INPUT" CDOWN>FILE NAME DESIRED FAST
COPY {LEFT 11>";F*
373
350 PRINT" {DOWN* WORKING "
360 OPEN 15,8, 15, "IO":GOSUB 480
370 OPEN 1,8.5, "O: "+F*+",P, W":GOSUB 480
380 :
390 REM STORE LOAD ADDRESS
400 PRINT#1 , CHR* < 1 ) CHR* (8) ;
410 :
420 REM STORE MAIN PROGRAM
430 FOR BLK=1 TO 20:NV=35:IF BLK=20 THEN
NV=4
440 FOR VL=1 TO NV:READ X:PRINT#1,CHR*(X
);:NEXT vl:next blk
450 :
460 CLOSE l: CLOSE 15: PRINT" { DOWN! ALL DON
E":END
470 :
480 INPUT#15,EN,EM*,ET,ES:IF EN<20 THEN
RETURN
490 PRINT" CDOWN> DISK ERROR": PRINT EN; EM*
set;es:close l: close is: end
500 :
510 REM START OF BLOCK #1
520 DATA 11, 8, 10, O, 158, 50, 48
530 DATA 54, 49, O, O, O, 32,204
540 DATA255, 32,231,255,160, 0,185
550 DATA215, 9, 32,210,255,200,192
560 DATA 21,208,245,160, O, 32,207
570 REM START OF BLOCK #2
580 DATA255,201, 13,240, 20,153, 60
590 DATA 3,200,208,243,160, 0,185
600 DATA236, 9, 32,210,255,200,192
610 DATA 21,208,245, 96,140, 52, 3
620 DATA192, 17,176,235,160, 0,185
630 REM START OF BLOCK #3
640 DATA 1, lO, 32,210,255,200,192
650 DATA 19,208,245, 32,207,255,141
660 DATA 53, 3,201, 83,240, 4,201
670 DATA 80,208,229,172, 52, 3,169
680 DATA 44,153, 60, 3,200,173, 53
690 REM START OF BLOCK #4
700 DATA 3,153, 60, 3,200,169, 44
710 DATA153, 60, 3,200,169, 82,153
720 DATA 60, 3,200,140, 52, 3, 32
730 DATA204,255, 32,170, 9, 32,192
740 DATA 9, 32, lOO, 9, 32,204,255
750 REM START OF BLOCK #5
760 DATA165, 251,240, 11,169, 1, 32
770 DATA195,255,169, 15, 32,195,255
780 DATA 96,169, 0,133,252,169, 11
790 DATA133,253, 160, O, 185, 42. lO
374
800 DATA 32,210,255,200,192* 20,208
81 O REM START OF BLOCK #6
820 DATA245, 162, 1, 32,198,255,160
830 DATA O, 32,207,255,145,252,165
840 DAT A 144,208, 7,200,208,244,230
B50 DATA253, 208,240, 132,252, 32,204
860 DATA255, 169, 1, 32,195,255,169
870 REM START OF BLOCK #7
880 DATA 15, 32,195,255,160, O, 185
890 DATA 62, 10, 32,210,255,200,192
900 DATA 49,208,245, 32,228,255,201
910 DATA 13,208,249, 32, 170, 9,172
920 DATA 52, 3,169, 87,153, 59, 3
930 REM START OF BLOCK #8
940 DATA 32,192, 9, 32,100, 9, 32
950 DATA204, 255, 165, 251 , 208, 142, 169
960 DATA 0,133,250,169, 11,133,251
970 DATA160, 0,185,111, 10, 32,210
980 DATA255, 200,192, 26,208,245,162
990 REM START OF BLOCK #9
lOOO DATA 1, 32,201,255,160. 0,177
1010 DATA250, 32,210,255,196,252,208
1020 DATA 6,165,251,197,253,240, 7
1 030 D AT A200 , 208 , 238 , 230 , 25 1 , 208 , 234
1040 DATA 32,204,255,169, 1, 32,195
1050 REM START OF BLOCK #10
1060 DATA255, 169, 15, 32,195,255,160
1070 DATA 0,185,137, 10, 32,210,255
1080 DATA200, 192, 21,208,245, 32,204
1090 DATA255, 32,228,255,201, 89,240
1100 DATA 8,201, 78,208,245, 32,204
1110 REM START OF BLOCK #11
1120 DATA255, 96, 76,215, 8, 32,204
1130 DATA255, 162, 15, 32,198,255, 32
1140 DAT A207, 255,201, 50,144, 39,133
1150 DATA251,160, 0,185, 20, 10, 32
1160 DATA210, 255, 200,192, 20,208,245
1170 REM START OF BLOCK #12
1180 DATA165,251, 32,210,255, 32,207
1190 DATA255,201, 13,240, 7, 32,210
1200 DATA255,208,244,240,242, 169, 1
1210 DATA133,251, 96, 32,207,255,201
1220 DATA 13,240, 4,208,247,240,245
1230 REM START OF BLOCK #13
1240 DATA169, 0,133,251, 96,165, 2
1250 DATA162, 40, 160, 10, 32,189,255
1260 DATA169, 15,162, 8,160, 15, 32
1270 DATA186,255, 32,192,255, 96,173
1280 DATA 52, 3,162, 60, 160, 3, 32
1290 REM START OF BLOCK #14
375
1300 DATA189,255, 169, 1,162, 8,160
1310 DATA 5, 32,186,255, 32,192,255
1320 DATA 96,142,147, 17, 70, 73, 76
1330 DATA 69, 78, 65, 77, 69, 32, 84
1340 DATA 79, 32, 67, 79, 80, 89, 63
1350 REM START OF BLOCK #15
1360 DATA 32, 13, 13, 18, 70, 73, 76
1370 DATA 69, 78, 65, 77, 69, 32, 84
1380 DATA 79, 79, 32, 76, 79, 78, 71
1390 DATA 13, 13, 13, 70, 73, 76, 69
1400 DATA 32, 84, 89, 80, 69, 32, 40
1410 REM START OF BLOCK #16
1420 DATA 83, 47, 80, 41, 63, 32, 13
1430 DATA 13, 18, 70, 65, 84, 65, 76
1440 DATA 32, 68, 73, 83, 75, 32, 69
1450 DATA 82, 82, 79, 82, 13, 73, 48
1460 DATA 13, 13, 82, 69, 65, 68, 73
1470 REM START OF BLOCK #17
1480 DATA 78, 71, 32, 68, 73, 83, 75
1490 DATA 32, 70, 73, 76, 69, 13, 13
1500 DATA 73, 78, 83, 69, 82, 84, 32
1510 DATA 79, 84, 72, 69, 82, 32, 68
1520 DATA 73, 83, 75, 69, 84, 84, 69
1530 REM START OF BLOCK #18
1540 DATA 13, 13, SO, 82, 69, 83, 83
1550 DATA 32, 18, 82, 69, 84, 85, 82
1560 DATA 78,146, 32, 87, 72, 69, 78
1570 DATA 32. 82, 69, 65, 68, 89, 13
1580 DATA145, 87, 82, 73, 84, 73, 78
1590 REM START OF BLOCK #19
1600 DATA 71, 32. 68, 73, 83, 75, 32
1610 DATA 70, 73, 76, 69, 32, 32, 32
1620 DATA 32, 32, 32, 13, 13, 65, 78
1630 DATA 79, 84, 72, 69, 82, 32, 67
1640 DATA 79, SO, 89, 32, 40, 89. 47
1650 REM START OF BLOCK #20
1660 DATA 78, 41, 63, 13
**********************
PROGRAM: EDIT T&S/C-64
**********************
Allows you to read or edit any block on a diskette. The display shows half a sector
at a time. You may toggle back and forth between the two halves. The left side of
the screen shows a hex dump of the sector. The right side shows the corresponding
ASCII characters. You may edit either the hex or the ASCII side. This is par-
ticularly useful when editing the diskette name or file names in the directory.
Since the sectors are read and written by bypassing the 1541's file management
routines, you may write out an edited block onto any position on any diskette.
NOTE: The disk ID's do not have to match.
376
Compatible with C-64 computer and 1541 disk drive only.
0 REM 1541 USERS' GUIDE APPENDIX E
1 REM COPYRIGHT: G. NEUFELD, 1984
2 :
3 REM EDIT ANY BLOCK ON A DISKETTE
4 :
lOO REM EDIT T&S C-64
110 POKE 53280, 1 1 : POKE 53281 ,11: P0KE650,
255
120 PRINTCHR*(14)CHR*(144) " CCLR* CSHIFT-E
>DIT CSHIFT-T>RACK & C SH I FT-S SECTOR <C>
<SHIFT-G3 . CSHIFT-NJEUFELD, 1984"
130 TR=18: SE=0: SX=16: FH=240: FL=15: C0=48:
C9=57: N2=2: N7=7: SS=1024: SC=55296: NC=40
140 D*="CH0ME3CD0WN 18V
1 50 B*= " " : ZC*=CHR* ( O ) : CR*=CHR* (13): TB=29
1 60 C I *=CHR* (148) +CHR* ( 34 ) +CHR* ( 20 )
170 CD*=CHR* ( 34 ) +CHR* ( 20 ) : QT*=CHR* ( 34 )
180 DIM HX(255),HX*(255)
190 DIM HT*(32),CH*(255)
200 GOSUB 3280: REM ZERO SID CHIP
210 FOR K=0 TO 63: CH* (K)=CHR* (K+Al) : CH* (
K+A2) =RV*+CH* (K) +R0*: NEXT
220 FOR K=64 TO 95: CH*(K)=CHR* (K+A2) :CH*
(K+A2) =RV*+CH* (K) +R0*: NEXT
230 FOR K=96 TO 127: CH* (K)=CHR* (K+Al ): CH
* (K+A2) =RV*+CH* (K) +R0*: NEXT
240 FOR K=0 TO 31 : Z=K*B:G0SUB2440:HT* (K
)="*"+z*+": ":next
250 B*=" ":nt=tr:goto 270
260 PRINT" CCLR>CSHIFT-E>DIT <SHIFT-T>RAC
K 8< {SHIFT-SJECTOR (C) CSHIFT-G>. {SHIFT
-N5EUFELD, 1984"
270 PRINT" {DOWN* <SHIFT-W>HICH TRACK "NT
280 PRINT" {UP} "TAB ( 12) ; : INPUT NT
290 IF NT<1 OR NT>35 THENPRINT" CUP> "TAB(
20) " {SHIFT-BJ AD TRACK" : GOSUB3540
300 IF NT<1 OR NT>35 THEN G0SUB2860: NT=T
R:GOTO 260
310 MS=20+2* (NT>17) + (NT>24> + (NT>30)
320 PRINT" {DOWN* CSHIFT-W>HICH SECTOR "S
E
330 PRINT"€UP>"TAB(13);: INPUT NS
340 IF NS<0 OR NS>MS THENPRINT" OJP}"TAB(
20 ) " { SH I F T-B> AD SECTOR " : G0SUB3540
350 IF NS<0 OR NS>MS THEN G0SUB2860 : GOTO
260
377
360 OPEN 15,8,15, "IO"
370 GOSUB 2500
380 OPEN 1,8,5. "#0"
390 GOSUB 2500
400 JOB=128: GOSUB 2930:REM SAME AS: PRIN
T#15, "Ul:5 0"TR;SE
410 IF GF=1 THEN CLOSE l:NT=TR:NS=SE: GOTO
260
420 PRINT" CDOWNJ {SHIFT-ONE MOMENT PLEA
SE . . . " : TR=NT : SE=NS
430 for k=0 to 255:get#1 , a*: z=asc(a*+zc*
>: gosub 2440: hx(K>=z:hx*<K)=z*: next
440 CLOSE 1
450 HF=0: GOSUB 2570
460 GOSUB 2700
470 GET A*: IF A*<>" "GOT0470
480 GET A*: IF A*="" GOTO 480
490 A=ASC(A*)-132
500 IF A<1 OR A>8 THEN GOSUB3540:GOTO 47
O
510 GOSUB 3440
520 ON A GOTO 570,1450,1730,2070,1040,15
10,1820,2210
530 :
540 REM EDIT <HEX)
550 G0SUB3540:G0T0 570:REM ERROR
560 GOSUB 3350
570 CC=PEEK (SS+N2+NC*ER+3*EC+EH)
580 POKE <SS+N2+NC*ER+3*EC+EH> , CCOR128
590 FOR K=l TO 30
600 GET A*: IF A*<>"" GOTO 670
610 NEXT
620 POKE < SS+N2+NC*ER+3*EC+EH ) , CCAND 1 27
630 FOR K=l TO 30
640 GET A*: IF A*<>"" GOTO 670
650 NEXT
660 GOTO 570
670 A=ASC(A*>
680 POKE ( SS+N2+NC*ER+3*EC+EH > , CC
690 IF A<133 OR A>140 GOTO 730
700 A=A- 132: GOSUB 3440
710 ON A GOTO 570,1450,1730,2070,1040,15
10,1820,2210
720 :
730 IF A=19 THEN EC=1:EH=0:ER=2:G0T0560
740 IF A=29 AND EH=0 THEN EH=l:GOTO 560
750 IF A=29 AND EH=1 AND EC<8 THEN EH=0:
EC=EC+l:GOTO 560
760 IF A=29 AND EH=1 AND EC=8 AND ER>=17
GOTO 550
378
770 IF A=29 AND EH=i AND EC=8 THEN EH=0:
EC=1 : ER=ER+1 : G0T0560
780 IF A=157 AND EH=I THEN EH=0:G0T0560
790 IF A=157 AND EH=0 AND EC>1 THEN EH=1
:EC=EC-l:GOTO 560
800 IF A=157 AND EH=0 AND EC=1 AND ER<=2
GOTO 550
810 IF A=157 AND EH=0 AND EC=1 THEN EH=1
:EC=8:ER=ER-l:GOTO 560
820 IF A=17 AND ER<17 THEN ER=ER+l:GOTO
560
830 IF A=145 AND ER>2 THEN ER=ER-I:G0T0
560
840 IF A=13 THEN EC=1:EH=0:IF ER<17 THEN
ER=ER+l:GOTO 560
850 IF<A<48)OR(A>57ANDA<65)OR(A>70> GOTO
550
860 POKE ( SS+N2+NC*ER+3*EC+EH) , AAND63
870 N=A-CO:IF A>C9 THEN N=N-N7
880 P=(ER-2)*8+EC-l: GOSUB 3350
890 C=HX <P+HF*128)
900 IF EH=0 THEN C=(C AND 15>+16*N
910 IF EH=1 THEN C=<C AND 240) +N
920 POKE <SS+TB+NC*ER+EC) . C
930 HX(P+HF*128)=C
940 Z=C:60SUB 2440:HX*<P+HF*128)=Z*
950 IF EH=0 THEN EH=1:G0T0 990
960 IF EH=1 AND EC<8 THEN EH=0:EC=EC+1:G
OTO 990
970 IF ER>=17 GOTO 990
980 ec=i:eh=o:er=er+i
990 GOTO 570
lOOO :
1010 REM EDIT (ASCII)
1020 GOSUB3540: GOTO 1040: REM ERROR
1030 GOSUB 3350: REM KEY CLICK
1 040 CC=PEEK ( SS+29+NC*ER+EC )
1 050 POKE ( SS+29+NC*ER+EC ) , CCOR 1 28
1060 FOR K=l TO 30
1070 GET A*: IF A*<>"" GOTO 1140
1080 NEXT
1 090 POKE ( SS+29+NC*ER+EC ) , CC AND 1 27
1100 FOR K=l TO 30
1110 GET A*: IF ASO"" GOTO 1140
1120 NEXT
1130 GOTO 1050
1140 A=ASC(A*)
1 150 POKE (SS+29+NC*ER+EC) , CC
1160 IF A<133 OR A>140 GOTO 1200
1170 A=A- 132: GOSUB 3440
379
1180 ON A SOTO 570,1450,1730,2070,1040,1
510,1820,2210
1190 :
1200 IF A=19 THEN EC=l:EH=0:ER=2:GOTO 10
30
1210 IF A=29 AND EC<8 THEN EC=EC+l:GOTO
1030
1220 IF A=29 AND EC=8 AND ER=17 SOTO 102
O
1230 IF A=29 AND EC=8 THEN EC=1: IFER<17T
HEN ER=ER+l:60T01030
1240 IF A=157 AND EC>1 THEN EC=EC-l:GOTO
1030
1250 IF A=157 AND EC=1 AND ER=2 GOTO 102
O
1260 IF A=157 AND EC=1 THEN EC=8: IFER>2T
HEN ER=ER-l:GOTO 1030
1270 IF A=17 AND ER<17 THEN ER=ER+1:G0T0
1030
1280 IF A=145 AND ER>2 THEN ER=ER-1:G0T0
1030
1290 IF A=13 THEN EC=1:IF ER<17 THEN ER=
ER+lZGOTO 1030
1300 IF(A<32)0R(A>95ANDA<160)0R(A>223) G
OTO 1020
1310 POKE ( SS+TB+NC*ER+EC ) , A
1320 GOSUB 3350
1330 CL=(A AND 15) +48: IF CL>C9 THEN CL=C
L+N7
1340 CH=(A AND 240) /SX+CO: IF CH>C9 THEN
CH=CH+N7
1 350 PR I NTLEFT* < D* , ER+ 1 ) T AB < 3*EC+2 ) CHR* (
CH)CHR*<CL)
1360 P=(ER-2)*8+EC-l
1370 HX(P+HF*128)=A
1380 Z=A: GOSUB 2440:HX*<P+HF*128)=Z*
1390 IF EC<8 THEN EC=EC+l:GOTO 1420
1400 IF EC=8 AND ER=17 GOTO 1040
1410 EC=l:IF ER<17 THEN ER=ER+1
1420 GOTO 1040
1430 :
1440 REM DISPLAY OTHER HALF
1450 IF HF=0 THEN HF=1:G0T0 1470
1460 HF=0
1470 GOSUB 2570
1480 GOTO 460
1490 :
1500 REM PRINT SECTOR
1510 GOSUB 2780
1520 OPEN 3,4
380
1530 PRINT#3:IF ST=0 BOTO 1590
1540 G0SUB3540
1550 PRINT" tSHIFT-DJEVICE #4 NOT PRESEN
T":CL0SE3
1560 INPUT" {DOWNJCSHIFT-OONTINUE <Y/N)
N CLEFT 3>";A*
1570 IF A*<>"Y" THEN GOSUB 27SO:GOSUB 27
00:G0T0470
1580 GOTO 1510
1590 GOSUB 2780
1600 INPUT" <SHIFT-I>S THE PAPER ADJUSTE
D (Y/N> YtLEFT 3>";A*
161 O IF A*<>"Y" THEN GOSUB 2780: GOSUB 27
O0:G0TO47O
1620 PRINT#3, " CSHIFT-TJRACK"MID*(STR*
<TR) , 2) " , {SHIFT-S>ECTOR"SE" <SHIFT-T>RAC
K ID="QT*ID*QT*
1630 PRINT#3
1640 FOR K=0 TO 248 STEP 8
1650 PRINT#3, " {DOWN} "HT* <K/8) HX* <K) HX* (K
+1 ) HX* <K+2> HX* <K+3> ;
1660 PRINT#3, HX* <K+4> HX* <K+5) HX* <K+6> HX*
(K+7);" "j
1670 FOR C=0 TO 7:PRINT#3„CH* (HX (K+C) > ; :
NEXT:PRINT#3
1680 NEXT:PRINT#3
1690 CLOSE 3: GOSUB 2780: GOSUB 2700
1700 GOTO 470
1710 :
1720 REM EXIT PROGRAM
1730 GOSUB 2780
1740 INPUT" {SHIFT-EJXIT PROGRAM NOW (Y/
N) N CLEFT 3J";A*
1750 IF A*<>"Y" G0T0460
1760 PRINT#15, "U; "
1770 CLOSE 1
1780 CLOSE 15
1790 GOSUB 2780: GOSUB 3280: END
1800 :
1810 REM WRITE SECTOR
1820 GOSUB 2780:NT=TR
1830 INPUT" {SHIFT-W>RITE SECTOR TO DISK
(Y/N> NtLEFT 3>";A*
1840 IF A*<>"Y" G0T0460
1850 GOSUB 2780:NS=SE: PRINT" €SHIFT-W>RI
TE TO TRACK "NT
I860 PRINT"«UP>"TAB(15);: INPUT NT
1870 IF NT<1 OR NT>35 THENPRINT"CUP>"TAB
<20>"{SHIFT-B>AD TRACK"
381
1880 IF NT<1 OR NT>35 THENG0SUB3540:G0SU
B 2860:NT=TR:G0T01830
1890 MS=20+2*(NT>17)+(NT>24)+(NT>30)
1900 PRINT" SECTOR "NS
1910 PRINT" <UP} "TAB (16);: INPUT NS
1920 IF NS<0 OR NS>MS THENPRINT" *UP}"TAB
< 20 ) " C SH I FT-B } AD SECTOR " : G0SUB3540
1930 IF NS<0 OR NS>MS THEN NS=SE:G0SUB28
60: GOTO 1830
1940 PR I NT "{DOWN* <SHIFT-0}NE MOMENT PLE
ASE "
1950 OPEN 1,8,5, "#0"
I960 PRINT#15, "B-P:5 O"
1970 FOR K=0 TO 255:PRINT#1 ,CHR*(HX (K) ) ;
ZNEXT
1980 J0B=144:G0SUB 2930:REM SAME AS: PRI
NT#15, "U2:5 0"TR;SE
1990 CLOSEl:IF GF=1 G0T0460
2000 tr=nt:se=ns
2010 print" {home} {rvs>cshift-t>rack"tr"
{left} , <shift-s>ector"se" {left} " ;
2020 if hf=0 then print" (first half) id
="CD*ID*
2030 IF HF=1 THEN PRINT" (LAST HALF) ID=
"CD* ID*
2040 GOTO 460
2050 :
2060 REM READ NEXT SECTOR
2070 GOSUB 2780
208O IF HX(0)=0 THEN PRINT" CSHIFT-T>HIS
SECTOR WAS THE LAST" :G0SUB3540
2090 IFHX(0)=0 THEN G0SUB2860:G0T0460
2100 PRINT" <SHIFT-N>EXT TRACK = "HX(O)
2110 IF HX(OXl OR HX(0)>35 THEN PRINT" i
UP} " TAB ( 20 ) " { SH I FT-B} AD TRACK " : G0SUB3540
2120 IF HX(OXl OR HX (O) >35 THEN GOSUB28
60: G0T0460
2130 PRINT" {SHIFT-NJEXT SECTOR ="HX(1)
2140 MS=20+2* (HX (O) >17) + (HX (O) >24) + (HX (O
)>30)
2150 IF HX(1)<0 OR HX(1)>MS THEN PRINT" €
UP> "TAB (20) " CSHIFT-B} AD SECTOR" : G0SUB354
O
2160 IF HXdXO OR HX(1)>MS THEN G0SUB28
60:G0T0460
2170 NT=HX(0)
21SO NS=HX(1)
2190 GOTO 2320
2200 :
2210 REM READ NEW SECTOR
382
2220 NT=TR
2230 GOSUB 2780: NS=SE: PRINT" {SHIFT-bOHI
CH TRACK "NT
2240 PRINT" OJP>"TAB( 12);: INPUT NT
2250 IF NT<1 OR NT>35 THEN PRINT" OJP> "TA
B ( 20 ) " <SH I FT-B* AD TRACK " : G0SUB3540
2260 IF NT<1 OR NT>35 THEN GOSUB2860:GOT
02220
2270 PRINT" {SHIFT-W>HICH SECTOR "NS
2280 MS=20+2* < NT > 1 7 ) + ( NT >24 ) + ( NT >30 )
2290 PRINT" OJP> "TAB (13);: INPUT NS
2300 IF NS<0 OR NS>MS THEN PRINT" tUPJ"TA
B <20) " CSHIFT-B3 AD SECTOR" : G0SUB3540
2310 IF NS<0 OR NS>MS THEN G0SUB2860 : GOT
02230
2320 OPEN 1,8,5, "#0"
2330 JOB= 128: GOSUB 2930: REM SAME AS: PR I
NT#15,"Ui:5 0"TR;SE
2340 IF GF=1 THEN CLOSE 1 : G0T0460
2350 tr=nt:se=ns
2360 PRINT" <DOWN> tSHIFT-0>NE MOMENT PLE
ASE. . . "
2370 PRINT#15, "B-P:5 O"
2380 FOR K=0 TO 255:GET#1 , A*: Z=ASC<A*+ZC
*) : GOSUB 2440: HX (K) =Z: HX* <K) =Z«: NEXT
2390 CLOSE 1
2400 HF=0: GOSUB 2570
24 lO GOTO 460
2420 :
2430 REM CONVERT TO HEX
2440 ZH=(FHANDZ)/SX+CO:lF ZH>C9 THEN ZH=
ZH+N7
2450 ZL=<FLANDZ)+CO:lF ZL>C9 THEN ZL=ZL+
N7
2460 Z*=CHR* < ZH) +CHR* ( ZL) +B*
2470 RETURN
2480 :
2490 REM READ DISK ERROR CHANNEL
2500 INPUT#15,EN,EM*,ET,ES
25 lO IF EN<19 THEN RETURN
2520 PRINT" {DOWNXSHIFT-DMSK ERROR:": GO
SUB3540
2530 PRINTEN;EM*;ET;ES
2540 CLOSE 1: CLOSE 15: END
2550 :
2560 REM DISPLAY SECTOR
2570 PRINT" <CLR> <!RVS>{SHIFT-T>RACK"TR" C
LEFT> , CSHIFT-S>ECTOR"SE" {LEFT! " ;
2580 IF HF=0 THEN PRINT" (FIRST HALF) ID
="CD*ID*
2590 IF HF=1 THEN PRINT" (LAST HALF) ID=
"CD* ID*
2600 PRINT
2610 FOR K=HF*128 TO HF* 128+ 120 STEP 8
2620 PRINTHT* <K/8) HX* <K) HX* (K+l ) HX* <K+2)
HX*(K+3>;
2630 PRINT HX*(K+4)HX*(K+5)HX*<K+6)HX*<K
+7)
2640 L=KAND 127: PS=SS+ 1 1 0+5*L : PC=SC+ 1 1 0+5
*L
2650 FOR C=0 TO 7: POKE PC+C,0:POKE PS+C,
HX<K+C):NEXT
2660 NEXT
2670 RETURN
26SO :
2690 REM DISPLAY COMMANDS
2700 PRINTD*"CDOWN> Fl = tSHIFT-E>DIT (H
EX) F2 = {SHIFT-EJDIT i CSHIFT-AJ CSHIF
T-S> {SHIFT-O CSHIFT-I 2> )
2710 PRINT" F3 = CSHIFT-0>THER HALF F
4 = CSHIFT-P2RINT SECTOR
2720 PRINT" F5 = £SHIFT-E>XIT PROGRAM F
6 = {SHIFT-W5RITE SECTOR
2730 PRINT" F7 = *SHIFT-N>EXT SECTOR F
8 = CSHIFT-NJEW SECTOR
2740 EC=l:EH=0:ER=2
2750 RETURN
2760 :
2770 REM CLEAR LOWER SCREEN
2780 PRINT D*
2790 FOR K=1T04
2800 PRINT"
2810 NEXT
2820 PRINT D*
2830 RETURN
2840 :
2850 rem wait for return
2860 print" {down* {shift-pjress {rvsj is
hift-r* cshift-ej {shift-t* {shift-uj {shift
-r}<:shift-n><:roff> to continue-
2870 set a*: if a*<>"" soto 2870
2880 bet a*: if a*ocr* goto 2880
2890 PRINT"<:UP>
■l
2900 RETURN
2910 :
2920 REM READ/WRITE SECTOR USING JOB QUE
UE
2930 TY=0
384
2940 PRINT#15, "M-W"CHR*(6)CHR*(0)CHR* (2)
CHR* (NT) CHR* (NS)
2950 PRINT#15, "M-W"CHR*(0)CHR*(0)CHR* ( 1 )
CHR* (176): REM SEEK TRACK
2960 PRINT#15, "M-R"CHR*(0)CHR*(0):REM WA
IT TILL JOB COMPLETE
2970 GET#15,A*:A=ASC(A*+ZC«)
2980 IF AM 27 GOTO 2960
2990 IF A=l GOTO 3060
3000 TY=TY+l:IF TY<5 GOTO 2940
30 lO GOSUB 2780: IF J0B=128 THEN PR I NT "{S
HIFT-R>EAD JOB ABORTED"
3020 IF J0B=144 THEN PRINT" CSHIFT-WJRITE
JOB ABORTED"
3030 PRINT"£DOWN> CSH I FT-T> ROUBLE ON THI
S TRACK"
3040 G0SUB3540: GOSUB 2860
3050 GF=l: RETURN
3060 PR I NT# 1 5 , " M-R " CHR* (18) CHR* ( O ) CHR* ( 2
>
3070 GET#15,A*:IF A*="" THEN A*=ZC*
3080 ID*=A*
3090 GET#15,A*:IF A*="" THEN A*=ZC*
3100 ID*=ID*+A*
31 lO TY=0
3120 PRINT#15, ,'M-W,,CHR*(6)CHR*(0)CHR*(2)
CHR*(NT)CHR*<NS)
3130 PRINT#15, "M-W"CHR*(0)CHR*(0)CHR*(1)
CHR* ( J OB ) : REM RE AD / WR I TE SECTOR
3140 PRINT#15, "M-R"CHR*(0)CHR*(0):REM WA
IT TILL JOB COMPLETE
3150 GET#15,A*:A=ASC(A*+ZC*)
3160 IF A>127 GOTO 3140
3170 IF A=l GOTO 3240
3180 TY=TY+l:IF TY<5 GOTO 3120
3190 GOSUB 2780:IF J0B=128 THEN PRINT"{S
HIFT-R>EAD JOB ABORTED"
3200 IF JOB=144 THEN PRINT" <SH I FT-W> RITE
JOB ABORTED"
3210 PRINT" CDOWN> <SH I FT-T> ROUBLE WITH T
HIS SECTOR"
3220 GOSUB3540: GOSUB 2860
3230 GF=l: RETURN
3240 PRINT#15, "B-P:5 O"
3250 GF=0: RETURN
3260 :
3270 REM SET-UP SID CHIP
3280 SID = 54272
3290 FOR K=SID TO SID+24
3300 POKE K,0
385
3310
3320
3330
3340
3350
3360
3370
3380
3390
3400
3410
3420
3430
3440
3450
3460
3470
3480
3490
3500
3510
3520
3530
3540
3550
3560
3570
3580
3590
3600
3610
NEXT
RETURN
REM DO KEY CLICK
POKE SID+13,240
POKE SID+24, 15
POKE SID+8,40
POKE SID+11,17
FOR K=l TO 30:NEXT
POKE SI D+ 11,0
RETURN
REM DO COMMAND TONE
POKE SID+1,20
POKE SID+5, 11
POKE SID+6,10
POKE SID+24, 15
POKE SID+4,21
FOR K=l TO 600:NEXT
POKE SID+4,0
RETURN
REM DO ERROR TONE
POKE SID+1,60
POKE SID+5, 11
POKE SID+6,10
POKE SID+24, 15
POKE SID+4,21
FOR K=l TO 750:NEXT
POKE SID+4,0
RETURN
it*********************
6. PROGRAM: MOD DISK NAME
***#**#*♦♦♦♦»*♦***#***
Allows you to modify the diskette name of any diskette. To use run program
insert diskette and edit.
Compatible with C-64, VIC-20, 4032 and 8032 computers, 1541, 2031 and 4040
disk drives.
0 REM 1541 USER'S GUIDE APPENDIX E
1 REM COPYRIGHT: G. NEUFELD, 1984
3 REM MODIFY A DISKETTE'S NAME
4 :
100 REM MOD DISK NAME
110 PRINT" CCLR} {DOWN} MODIFY DISKETTE NA
ME"
120 PRINT" {DOWN* INSERT DISKETTE TO BE M
ODIFIED"
386
130 PR I NT "{DOWN* PRESS RETURN WHEN READY
ll
140 SET A*: IF A*OCHR*<13> GOTO 140
150 PRINT" {UP} WORKING
160 OPEN 15,8, 15," 10"
170 GOSUB 470
180 OPEN 1,8,5, "#"
190 PRINT#15, "Ul:5 O 18 O"
200 PRINT#15,"B-P:5 144"
210 DN*=""
220 FOR K=l TO 16
230 GET#1,A*:IF A*="" THEN A*=CHR*(0>
240 DN*=DN*+A*
250 NEXT
260 PRINT" {CLR} {DOWN} EDIT DISK NAME"
270 PRINT" £DOWN 21 "DN*
280 PRINT" "
290 PRINT" CUP 2> ";: INPUT NN*
300 IF LEN<NN*><17 GOTO 360
310 IF RIGHT* <NN*,1>=CHR*<160> THEN NN*=
LEFT* ( NN* , LEN < NN* ) - 1 ) : GOTO 300
PRINT" {DOWN} ** NAME TOO LONG **"
PRINT" {DOWN} PRESS RETURN TO GO ON"
GET A*: IF A*OCHR*<13) GOTO 340
GOTO 260
CHANGE ON DISKETTE (Y
INPUT" {DOWN 2}
N{LEFT 3>";X*
IF ASC(X*><>89
THEN CLOSE 1: CLOSE 15
320
330
340
350
360
/N)
370
:END
380 IF LEN<NN*)<16 THEN NN*=NN*+CHR* ( 1 60
):GOTO 380
390 PRINT#15, "B-P:5 144"
PRINT#1,NN*;
PRINT#15,"U2:5 O 18 0"
GOSUB 470
PRINT#15,"I0"
CLOSE l: CLOSE 15
PRINT" CDOWNJ DONE" : END
400
410
420
430
440
450
460
470
480
490
500
510
INPUT#15. EN, EM*, ET, ES
IF EN<20 THEN RETURN
PRINT" {DOWN} DISK ERROR"
PR I NTEN ; EM* ; ET ; ES
CLOSE l: CLOSE 15: END
**********************
7. PROGRAM: FIX DIRECTORY
**********************
387
Provides automated recovery of files lost due to a damaged directory sector or
inadvertent issuance of a "short NEW" command.
To recover files after a "short NEW" command specify "1", the sector to be
recovered.
To recover files lost because of a read error on track 18, you must know which
sector is causing the problem. Use this short program to scan track 18 for this.
100 REM FIND ERRORS ON TRACK 18
110 PRINT" €CLR>FIND ERRORS IN THE DIRECT
ORY"
120 PRINT" <DOWN> INSERT DISKETTE"
130 PRINT" «DOWN> PRESS €RVS>RETURN<ROFF>
TO CONTINUE"
140 GET A*: IF A*OCHR*<13> GOTO 140
150 PRINT" <CLR>ERROR ANALYSIS OF TRACK 1
8 {DOWN}"
160 OPEN 15,8,15, "10"
170 OPEN 1,8,5, "#"
180 DIM SE<20)
190 se<i)=o:kt=i:ne=o
200 FOR K=2 TO 19
210 SE<K)=KT
220 KT=KT+3:IF KT>18 THEN KT=KT-17
230 NEXT
240 FOR K=l TO 19
250 PRINT#15, "Ul:5 O 18"SE(K)
260 INPUT#15,EN,EM*,ET,ES*
270 IF EN<20 GOTO 300
280 NE=NE+1
290 PR I NT "SECTOR #"SE <K) "ERROR #"EN" - "
EM*
300 NEXT
310 CLOSE l: CLOSE 15
320 IF NE=0 THEN PRINT" (DOWN! NO ERRORS F
OUND ON TRACK 18"
Once you have found the problem sector, write down the sector number. Load and
run the FIX BAD DIR program and specify this sector as the one to be recovered.
Compatible with C-64 and VIC-20 computers and the 1541 disk drive.
100 REM DIR-FIX FOR 1541 AND C-64/VIC-20
110 PRINT" CCLR} <RVS> RECOVER DIRECT
ORY SECTOR
120 PRINT" {DOWN* THIS UTILITY WILL RECOVE
R A SECTOR"
130 PR I NT "OF THE DIRECTORY. TO RECOVER A
FTER A"
140 PR I NT "SHORT NEW ENTER * 1 * AS THE SEC
TOR TO"
150 PR I NT "BE RECOVERED. NOTE THAT YOU CA
NNOT"
160 PR I NT "RECOVER ANY FILES AFTER A FULL
NEW. "
170 PRINT" CDOWN> INSERT DISKETTE WITH BAD
DIRECTORY"
180 PRINT" <DOWN>PRESS CRVS> RETURN CROFF>
TO CONTINUE"
190 GET A*: IF A*<>"" GOTO 190
200 GET A*: IF A*OCHR*(13) GOTO 200
210 INPUT" {UP>WHICH SECTOR TO RECOVER 1
{LEFT 3>";A*
220 DX=INT(VAL<A*>>
230 IF DX>0 AND DX<19 GOTO 270
240 PRINT"CDOWN>I DON'T DO SECTOR #"DX
250 GOTO 210
260 :
270 Z*=CHR*<0)
280 DIM TL(200),SL(200),BM(35,24),B'/.(35,
2>,BF<35>
290 BO= 1 : B 1 =2 : B2=4 : B3=8 : B4= 1 6 : B5=32 : B6=6
4:B7=128
300 OPEN 15,8, 15, "IO"
310 GOSUB 319C»:IF FZ=1 THEN STOP
320 OPEN 1,8,5, "#"
330 GOSUB 3190: IF FZ=1 THEN STOP
340 :
350 IF DX=1 GOTO 870
360 :
370 REM NOT SECTOR 18/1, SO CORRECT ANY
BAD LINKS
380 REM AND UNSCRATCH ANY SCRATCHED FILE
S IN THE
390 REM PRECEEDING DIRECTORY BLOCKS
400 PRINT" {DOWN* CHECKING FOR CORRECT LIN
KS AND"
410 PR I NT "SCRATCHED FILES IN PREVIOUS BL
OCKS"
420 ZD=1
430 PRINT#15, "Ul:5 O 18"ZD
440 PR I NT "CHECKING SECTOR 18, "ZD
450 GOSUB 3190: IF FZ=1 THEN STOP
460 :
470 REM CHECK LINK TO NEXT SECTOR
480 PRINT#15, "B-P:5 O"
490 GET#1 , A*: TL=ASC < A*+Z*>
389
500 GET#1,A*:SL=ASC(A*+Z*>
510 IF TL=18 AND SL>1 AND SL<19 GOTO 630
520 :
530 REM CORRECT BAD LINK
540 SL=ZD+3:TL=18
550 IF SL>18 THEN SL=SL-17
560 PRINT#15„ "B-P:5 O"
570 PRINT#1 , CHR* <TL> CHR* <SL> ?
580 PRINT#15.,"U2:5 O 18"ZD
590 PRINT"FIXING LINK ON 18/"ZD
600 GOSUB 3190:IF FZ=1 THEN STOP
610 :
620 REM SCAN FOR SCRATCHED FILES
630 WF=0:FOR K=0 TO 7
640 PRINT#15,'*B-P:5,,2+K*32
650 GET#1„A*:FT=ASC<A*+Z*>
660 IF FT > 128 GOTO 770
670 :
680 REM UNSCRATCH FILE ENTRY AND CREATE
690 REM DUMMY ENTRY IF NECESSARY
700 GET#1 , A*: ZT=ASC ( A*+Z*>
710 GET#1,A*:ZS=ASC<A*+Z*>
720 WF=l:NT=FT OR 128
730 PR I NT# 15," B-P : 5 " 2+K*32
740 PRINT#1 , CHR* <NT) ;
750 IF ZT=0 OR ZT>35 THEN PRINT#1 .CHR* ( 1
7)CHR*<0>"- DUMMY ENTRY - "5
760 IF FT>0 THEN PRINT#1 , CHR*< 17) CHR* <0>
"- DUMMY ENTRY - ";
770 NEXT K
780 :
790 REM REWRITE SECTOR IF ANY CHANGES
800 IF WF=1 THEN PRINT#15. "U2:5 O 18"ZD
810 GOSUB 3190: IF FZ=1 THEN STOP
820 :
830 ZD=SL
840 IF ZDODX GOTO 430
850 :
860 REM COMPUTE FILE NUMBERS FOR NEW FIL
ES
870 NF=0
880 IF (DX-2)/3=INT((DX-2)/3) THEN NF=48
890 IF DX/3=INT<DX/3> THEN NF=96
900 NF=NF+8* INT<(DX-l)/3)
910 DY=DX+3:IF DY>18 THEN DY=DY-17:REM D
Y=NEXT DIR SECTOR
920 PRINT" {DOWN} STARTING WORK ON TRACK 1
8, SECTOR"DX
390
930 :
940 REM CHECK FOR READ ERROR ON SECTOR
950 PRINT#15, "Ul:5 O 18"DX
960 INPUT#15,EN,EM*,|ET,ES
970 IF EN<19 GOTO 1650
980 PR I NT "{DOWN* DISK ERROR WHEN READING
SECTOR"
990 print" "EN;em*;et;es
iooo :
1010 IF EN<22 OR EN>24 GOTO 1080
1020 :
1030 PRINT" CDOWN>NOTE: MOST 22, 23, AND
24 READ ERRORS"
104O PRINT" CAN BE CORRECTED BY NUL
LING OUT"
1050 PRINT" THE SECTOR AND THEN REC
OVERING. "
1060 GOTO 1130
1070 :
1080 PRINT" {DOWN} NOTE: 20. 21. AND 27 RE
AD ERRORS ARE"
1090 PRINT" DUE TO ERRORS IN THE HE
ADER AND"
llOO PRINT" SO ARE NOT USUALLY CURE
D IF YOU"
lllO PRINT" NULL OUT THE SECTOR."
1120 :
1130 PRINT" {DOWN} YOUR OPTIONS ARE:"
1140 PRINT" CDOWN} 1. NULL OUT THIS SECTOR
II
1150 PRINT "2. BRIDGE OVER THIS SECTOR"
1160 PR I NT "3. TERMINATE THE PROGRAM"
1170 INPUT" {DOWN* YOUR CHOICE IS <l-3) 1
{LEFT 3>";A*
1180 A=ASC(A*)-4S
1190 IF A<1 OR A>3 THEN PR I NT "ENTER ONLY
1, 2, OR 3": GOTO 1170
1200 :
1210 IF A=3 THEN CL0SE1:CL0SE15: END
1220 IF A=l GOTO 1570
1230 :
1240 REM BRIDGE OVER TROUBLED SECTOR
1250 IF DX=1 THEN PRINT" {DOWN} YOU CAN'T
BRIDGE OVER 18/1": GOTO 1140
1260 DW=DX-3:IF DW<1 THEN DW=DW+18:REM P
REVIOUS SECTOR
1270 PRINT#15, "Ul:5 O 18"DW
1280 GOSUB 3190: IF FZ=1 THEN STOP
1290 PRINT#15, "B-P:5 O"
1300 PRINT#1.CHR*<18)CHR*(DY);
391
1310 PRINT#15, "U2:5 O iS"DW
J320 GOSUB 3190: IF FZ=1 THEN STOP
1330 :
1340 PRINT" {DOWN* SECTOR 18."DX"NO LONGER
IN DIRECTORY"
1350 PRINT" {DOWN} VALIDATING DISKETTE TO
CORRECT BAM"
1360 PRINT#15, "VO"
1370 GOSUB 3190: IF FZ=1 THEN STOP
13SO CLOSE l:OPEN 1,8, 5, "tt"
1390 GOSUB 3190: IF FZ=1 THEN STOP
1400 GOSUB 4490
1410 :
1420 REM FIND LAST BLOCK IN DIRECTORY FI
LE
1430 ZD=1
1440 PRINT#15, "Ul:5 O 18"ZD
1450 GOSUB 3190: IF FZ=1 THEN STOP
1460 PRINT#15,"B-P:5 O"
1 470 GET# 1 , A* : Z T=ASC < A*+Z* >
1480 GET#1 , A*: ZS=ASC <A*+Z*>
1490 IF ZT=18 AND ZS<19 AND ZS>0 THEN ZD
=ZS:GOTO 1440
1500 IF ZTOO OR ZS0255 THEN PRINT" CDOW
N>BAD DIR LINK ON SECTOR "ZD: GOTO 1540
1510 ZD=ZD+3: IF ZD>19 THEN ZD=ZD-18
1520 PRINT" <DOWN> TO RECOVER LOST FILES R
UN PROGRAM AGAIN"
1530 PR I NT "AND SPECIFY RECOVERY OF SECTO
R #"ZD
1 540 CLOSE 1 : CLOSE 1 5 : END
1550 :
1560 REM NULL OUT THE SECTOR AND TRY AGA
IN
1570 PRINT#15, "B-P:5 0"
1580 PRINT*! , Z*; CHR* (255) ;
1590 FOR K=2 TO 255: PRINT#1, Z*; : NEXT
1600 PRINT#15, "U2:5 O 18"DX
1610 GOSUB 3190:IF FZ=1 THEN STOP
1620 GOTO 950
1630 :
1640 REM CHECK IF SECTOR IS ALL NULLS
1650 PRINT#15, "B-P:5 2"
1660 M*= "DIRECTORY SECTOR NOT BLANK"
1670 FOR K=2 TO 255
1680 GET#1,A*:IF A*<>"" THEN K=255:NEXT:
PR I NT " {DOWN* " M* : GOTO 1 1 30
1690 NEXT
1700 :
1710 REM CHECK IF FOLLOWING SECTOR HAS E
NTRIES
392
1720 PRINT#15, "Ui:5 0 18"DY
1730 GOSUB 3190: IF FZ=1 THEN STOP
1740 PRINT#15, "B-P:5 O"
1750 GET#1,A*:A=ASC<A*+Z*>
1760 get#i,a*:b=asc<a*+z*>
1770 DF=0
1780 IF A=0 AND B=0 GOTO 1910: REM VIRGIN
SECTOR
1790 IF A=75 AND B=l GOTO 1910: REM VIRGI
N SECTOR
1SOO :
1810 REM RELINK DIRECTORY TO NEXT SECTOR
1820 PRINT#15, "Ul:5 O 18"DX
1830 GOSUB 3190: IF FZ=1 THEN STOP
1840 PRINT#15, "B-P:5 O"
1850 PRINT#1,CHR*<18)CHR*(DY>;
I860 PRINT#15, "U2:5 O 18"DX
1870 GOSUB 3190:IF FZ=1 THEN STOP
1880 PRINT" CDOWN* VALIDATING DISKETTE TO
RECOVER ALL"
1890 PRINT"FILES EXCEPT THE 8 DAMAGED EN
TRIES. "
1900 GOTO 1920
1910 PRINT" CDOWNJVALIDATING BAM FOR STAR
T OF DIRECTORY"
1920 CLOSE 1
1930 PRINT#15, "VO":REM ALLOCATE RECOVERE
D FILES
1940 GOSUB 3190: IF FZ=1 THEN STOP
1950 GOSUB 4490
1960 OPEN 1,8,5, "#"
1970 :
1980 REM READ IN THE BAM
1990 PRINT" {DOWN} READING IN BAM. 15 SECO
ND PAUSE..."
2000 GOSUB 3600
2010 :
2020 REM RE-CREATE FIRST EIGHT ENTRIES
2030 MN=17:MX=19
2040 FOR F=0 TO 7:Z1=0
2050 PRINT#15, "U;'*:REM RESET 1541 DRIVE
2060 CLOSE l:REM CLOSE FILE IN COMPUTER
2070 PRINT#15, "10"
2080 PRINT" {DOWN} WORKING ON FILE ENTRY #
"NF+F+1
2090 REM OPEN NEW FILE TO FOR NEW ENTRY
2100 OPEN 2,8,4, "FILE #"+STR* (NF+F+1 >+",
P,W"
2110 INPUT#15,E,E*:IF E<20 GOTO 2170
2120 Zl=Zl+l:IF ZK5 THEN CLOSE 2:GOTO 2
050
2130 PRINT" {DOWN* JOB ABORTED" : PR I NTE;E«:
CLOSE2: CLOSE 15: STOP
2140 :
2150 REM NOTE: BAM IS NOW OVER ALLOCATED
2160 REM FORCE NEW ENTRY TO BE UNCLOSED
FILE
2 1 70 PR I NT# 15," M-W " CHR* ( 43+4 ) CHR* < 2 ) CHR*
<1>CHR*<255>
2 ISO CLOSE 2: CLOSE 15: REM IN COMPUTER ON
LY
2190 :
2200 REM REOPEN FILES AND GRAB NEW ENTRY
2210 OPEN 15,8, 15, "10"
2220 OPEN 1,8,5, "#"
2230 PRINT#15, "Ul:5 O 18"DX
2240 GOSUB 3190: IF FZ=1 THEN STOP
2250 PR I NT# 1 5 , " B-P : 5 " 3+32*F
2260 GET# 1 , A* : TL=ASC < A*+Z* >
2270 IF TL<MN THEN MN=TL
22SO IF TL>MX THEN MX=TL
2290 GET# 1 , A* : SL=ASC ( A*+Z* >
2300 GOSUB 3320:REM TRACE NEW ENTRY FILE
CHAIN
23 lO REM SET UP LENGTH OF NEW ENTRY
2320 BH=INT<NB/256):BL=NB-256*BH
2330 PRINT#15, "Ul:5 O 18"DX
2340 GOSUB 3190: IF FZ=1 THEN STOP
2350 PR I NT# 1 5 , " B-P : 5 " 30+32*F
2360 PRINT#1 , CHR* <BL) CHR* <BH) ;
2370 PRINT#15, "U2:5 O 18"DX
2380 GOSUB 3190: IF FZ=1 THEN STOP
2390 IF FL=1 GOTO 2050: REM BAD FILE
2400 IF NB>6 GOTO 2430:REM CAN'T BE A SI
DE SECTOR FILE
2410 GOSUB 3940: REM CHECK IF THIS IS A S
IDE SECTOR FILE
2420 IF RF=1 THEN F=F-1:RF=0:G0T0 2770
2430 GOSUB 3810:REM ALLOCATE NEW FILE IN
BAM
2440 REM GET LOAD ADDRESS OF THE NEW FIL
E
2450 PRINT#15,"Ul:5 0"TL(1)SL(1)
2460 GOSUB 3190: IF FZ=1 THEN STOP
2470 PRINT#15, "B-P:5 2"
2480 GET# 1 , A* : A=ASC < A*+Z * )
2490 GET#1,A*:B=ASC(A*+Z*>
2500 LA=256*B+A
394
25 1 0 L A*= " ( " +M I D* ( STR* < 256*B+ A ) , 2 > + " ) "
2520 H=<240ANDB>/ 16+48: IF H>57 THEN H=H+
7
2530 REM CONVERT LOAD ADDRESS TO HEX NOT
AT I ON
2540 LH*="*"+CHR*<H>
2550 H=<15ANDB)+48:IF H>57 THEN H=H+7
2560 LH*=LH*+CHR*<H>
2570 H=(240ANDA>/ 16+48: IF H>57 THEN H=H+
7
2580 LH*=LH*+CHR*(H>
2590 H=(15ANDA)+48:IF H>57 THEN H=H+7
2600 LH*=LH*+CHR*<H)
2610 PRINT" LOAD ADDRESS OF FILE = "LH*"
"LA*
2620 FT=l:FT(F)=l
2630 :
2640 IF LA=1025 OR LA=2049 OR LA=4097 TH
EN FT=2:FT<F>=2
2650 IF LA=4097 OR LA=4609 OR LA= 16270 T
HEN FT=2:FT(F)=2
2660 IF LA=32768 OR LA=36864 OR LA=40960
THEN FT=2:FT(F>=2
2670 IF FT=1 THEN PRINT" ASSUMED TO BE A
SEQUENTIAL FILE"
2680 IF FT=2 THEN PRINT" ASSUMED TO BE A
PROGRAM FILE"
2690 REM PUT FILE TYPE IN DIRECTORY ENTR
Y
2700 PRINT#15,"Ul:5 O 18"DX
2710 GOSUB 3190:IF FZ=1 THEN STOP
2720 PR I NT# 1 5 , " B-P : 5 " 2+32*F
2730 PRINT#1,CHR*(128+FT>;
2740 PRINT#15,"U2:5 O 18"DX
2750 GOSUB 3190: IF FZ=1 THEN STOP
2760 :
2770 NEXT F
2780 :
2790 REM SEE IF WE HAVE ANY SEQ FILES AN
D NEED
2800 REM TO CHECK FOR POSSIBLE SIDE SECT
OR FILES
2810 FOR K=0 TO 7: IF FT(K)=1 THEN K=7:NE
XT: GOTO 2860
2820 NEXT
2830 GOTO 3040: REM DON'T BOTHER CHECK IN
G
2840 :
2850 REM CHECK FOR SIDE SECTOR FILES
2860 PRINT" <DOWN>CHECKING FOR ANY SIDE S
ECTOR FILES"
395
2870 FOR TX=17 TO MN STEP-1
2880 FOR SX=0 TO 20
2890 IF BM(TX,SX>=0 GOTO 2940
2900 TL<1)=TX
2910 SL<1>=SX
2920 NB=1
2930 GOSUB 3940
2940 NEXT SX:NEXT TX
2950 FOR TX=19 TO MX
2960 FOR SX=0 TO 20
2970 IF BM(TX,SX)=0 GOTO 3020
29SO TL(1>=TX
2990 SL(1)=SX
3000 NB=1
3010 GOSUB 3940
3020 NEXT SXZNEXT TX
3030 :
3040 CLOSE 1
3050 PRINT" CDOWNJSCRATCHING ANY DUMMY EN
TRIES CREATED"
3060 PRINT#15, "SO:- DUMMY ENTRY -*"
3070 PRINT" <DOWN>DOING FINAL VALIDATION
TO UPDATE BAM"
3080 PRINT#15, "VO"
I NPUT# 1 5 , EN , EM* , ET f ES
IF EN<20 THEN PRINT" {DOWN* JOB COMPL
: CLOSE 15: END
PR I NT "FINAL VALIDATION ABORTED, BAM
INCORRECT"
3120 PRINT" {DOWN* "EN; EM*; ET;ES
3130 PR I NT "COPY ALL FILES TO ANOTHER DIS
KETTE.":END
3140 :
REM
REM SUBROUTINES
REM
REM CHECK DISK STATUS
INPUT#15,EN,EM*,ET,ES
IF EN<20 THEN FZ=0: RETURN
print" cdownj job aborted"
print" hen;em*;et;es
CLOSE l: CLOSE 15
FZ=l: RETURN
3250 :
3260 REM FOLLOW FILE,
CK BAM
3270 REM ON ENTRY
SECTOR LINK
3280 REM ON EXIT:
3290 REM
T & S LINKS
3090
3100
ETED
3HO
3150
3160
3170
3180
3190
3200
32 lO
3220
3230
3240
TL
STORE LINKS, & CHE
= TRACK LINKS SL =
NB m NUMBER OF BLOCKS
TLO JtSLO = LIST OF
3300 REM FL = O IF ALL BLOCKS F
REE
3310 REM FL = 1 IF NOT ALL BLOC
KS FREE
3320 NB=1
3330 IF BM<TL,SL>=0 THEN M*=" FIRST BLOCK
IN USE": GOTO 3510
3340 TL < NB ) =TL : SL ( NB ) =SL
3350 PRINT#15, "Ul:5 0"TL;SL
3360 GOSUB 3190: IF FZ=1 THEN STOP
3370 PRINT#15, "B-P:5 O"
3380 GET#1,A*:TL=ASC<A*+Z*>
3390 GET#1,A*:SL=ASC<A*+Z*>
3400 IF TL=0 AND SL>0 GOTO 3480: REM LAST
BLOCK
3410 IF TL=0 AND SL=0 THEN M*=" VIRGIN SE
CTOR":GOTO 3510
3420 IF TL=75 AND SL=1 THEN M*=" VIRGIN S
ECTOR": GOTO 3510
3430 IF TL>35 THEN M*="BAD LINK": GOTO 35
10
3440 IF BM<TL,SL)=0 THEN M*="IN USE": GOT
O 351 0
3450 NB=NB+1
3460 TL < NB > =TL : SL < NB ) =SL
3470 GOTO 3350
3480 PRINT" FILE W I TH"NB" BLOCKS TRACED O
K"
3490 FL=0
3500 RETURN
3510 PRINT" TROUBLE: "M*
3520 PRINT" AT TRACK " TL < NB ) " , SECTOR "SL(
NB)
3530 ZZ=ZZ+l:IF ZZ>5 THEN CLOSE 15:PRINT
"JOB ABORTED!"
3540 PRINT" {DOWN* VALIDATING AGAIN"
3550 PRINT#15,"VO": CLOSE l:OPEN 1,8,5, "#
3560 FL=1
3570 RETURN
"^58("> :
3590 REM READ IN & STORE THE BAM
3600 PRINT#15. "Ul:5 O 18 O"
3610 GOSUB 3190: IF FZ=1 THEN STOP
3620 PRINT#15, "B-P:5"4
3630 FOR TR=1 TO 35
3640 GET# 1 , A* : BF ( TR > =ASC < A*+Z* )
3650 FOR K=0 TO 2
3660 GET* 1 , A* : A=ASC < A*+Z* ) : K8=8*K
3670 B7.(TR,K)=A
3680 IF A AND BO THEN BM(TR,K8)=1
397
3690 IF A AND
3700 IF A AND
3710 IF A AND
3720 IF A AND
3730 IF A AND
3740 IF A AND
3750 IF A AND
3760 NEXT K
3770 NEXT TR
3780 RETURN
3790 :
3800
3810
3820
3830
3840
3850
Bl THEN BMtTR, 1+K8)=1
B2 THEN BM<TR,2+K8)=1
B3 THEN BM(TR,3+K8)=1
B4 THEN BM(TR,4+K8)=1
B5 THEN BM(TR,5+K8)=1
B6 THEN BM<TR,6+K8)=1
B7 THEN BM(TR, 7+K8)=l
IN BAN
REM ALLOCATE FILE
FOR K=l TO NB
BM<TL<K),SL(K))=0
BY=INT<SL<K>/8)
BI=SL(K)-8*BY
BY. <TL (K> , BY) =B% <TL <K) , BY)
AND <255-
2 BI)
3860 BF<TL(K))=BF(TL(K))-1
IF TL<KXMN THEN MN=TL<K)
IF TL<K)>MX THEN MX=TL(K)
NEXT
GOSUB 4610: REM PUT GOOD BAM TO DISK
3870
3880
3890
3900
ETTE
3910
3920
3930
3940
3950
3960
3970
3980
3990
4000
40 lO
4020
RETURN
IF REL FILE
REM CHECK
RF=1
PRINT#15, "Ul:5 0"TL<1)SL<1)
GOSUB 3190: IF FZ=1 THEN STOP
PRINT#15, "B-P:5 4"
GET#1 , A*: TL=ASC < A*+Z«)
IF TLOTL(l) THEN RF=0: RETURN
GET#1 , A*: SL=ASC < A*+Z*)
IF SLOSL(l) THEN RF=0: RETURN
PRINT"<:DOWNJ FOUND WHAT MAY BE PART
OF A REL FILE"
4030 INPUT" IS THIS POSSIBLE (Y/N) NCLE
FT 3>";A*
4040 IF (ASC (A*) AND 127)089 THEN RF=0:RE
TURN
4050 PRINT#15, "B-P:5 3"
4060 GET# 1 , A* : RS=ASC ( A«-t-Z* )
4070 PRINT#15, "B-P:5 16"
4080 GET# 1 , A* : FT=ASC ( A*+Z* )
4090 GET#15A*:FS=ASC<A*+Z*)
4100 PRINT#15, "Ul:5 O 18"DX
4110 GOSUB 3190:IF FZ=1 THEN STOP
4120 FOR KX=0 TO F-l
4130 PRINT#15, "B-P:5"3+32*KX
4 1 40 GET# 1 , A* : A=ASC < A*+Z* )
398
4150 IF AOFT GOTO 4440
4160 GET#1,A*:B=ASC(A*+Z*)
4170 IF BOFS SOTO 4440
4180 PRINT" *D0WN> SIDE SECTOR FILE FOR F
ILE #"KX+1
4190 PRINT" FILE #"KX+1 "CONVERTED TO REL
FILE"
4200 PRINT" RECORD SIZE I S"RS" CHARACTERS
ll
4210 FT(KX)=4
4220 PRINT#15, "B-P:5"2+32*KX
4230 PRINT#1, CHR* (128+4);: REM REL FILE T
YPE
4240 PRINT#15, "B-P:5"21+32*KX
4250 PRINT#1 , CHR* <TL) CHR* (SL) CHR* <SS> ; : R
EM SIDE SECTOR LINK
4260 PRINT#15, "B-P:5"30+32*KX
4270 GET#1,,A*:BL=ASC(A*+Z*)
4280 GET# 1 , A* : BH=ASC ( A*+Z* )
4290 PRINT#15, "U2:5 0 18"DX
4300 GOSUB 3190:IF FZ=1 THEN STOP
43 lO GOSUB 3320: REM TRACE SIDE SECTOR CH
AIN
4320 IF FL=0 THEN GOSUB 3810: REM ALLOCAT
E IT
4330 PRINT#15, "Ul:5 O 18"DX
4340 BL=BL+NB
4350 IF BL>255 THEN BL=BL-256:BH=BH+1
4360 PR I NT# 15," B-P : 5 " 30+32*KX
4370 PR I NT# 1 , CHR* < BL ) CHR* ( BH ) ;
4380 KX=F-l:NEXT:IF F>7 GOTO 4410
4390 PR I NT# 1 5 , " B-P : 5 " 2+32*F
4400 PRINT#1,CHR*(0)CHR*(0)CHR*(0) ;
4410 PRINT#15, "U2:5 O 18"DX
4420 GOSUB 3190: IF FZ=1 THEN STOP
4430 RETURN
4440 NEXT KX
4450 PRINT" DOESN'T FIT WITH ANY OTHER F
ILE"
4460 RF=0: RETURN
4470 :
4480 REM FORCE WRITE OF BAM TO DISKETTE
4490 TZ=0
4500 PRINT#15, "M-WCHR* ( 14) CHR* <0) CHR* (2
) CHR* (18) CHR* CO)
4510 PRINT#15, "M-W"CHR* <4) CHR* <0) CHR* ( 1 )
CHR* (144)
4520 PRINT#15, "M-R"CHR*(4)CHR*(0)
4530 GET#15,A*:A=ASC(A*+CHR*(0))
4540 IF A>127 GOTO 4520
4550 IF A=l THEN RETURN
399
4560 TZ=TZ+l:IF TZ<10 GOTO 4500
4570 PRINT-JOB ABORTED, PROBLEMS WITH 18
/O"
4580 CLOSE 1: CLOSE 15: STOP
4590 :
4600 REM STORE CORRECT BAM IN DRIVE RAM
4610 FOR ZT=1 TO 35
4620 M*=CHR* (BF (ZT) ) +CHR* <B% (ZT, O) ) +CHR*
(BX(ZT,1))+CHR*<B%(ZT,2>)
4630 PRINT#15, "M-W"CHR*(4*ZT)CHR*(7)CHR*
(4)M*
4640 NEXT
4650 GOSUB 4490
4660 RETURN
8. PROGRAM: THE UNSCRATCHER
ft***********************
This program recovers any scratched file. It displays a list of all scratched files on
the diskette and you may choose which one to unscratch. The file is traced to
ensure that it does not contain any read errors and that it has not been overwritten.
The program works with any type of file, including relative files.
Compatible with C-64, VIC-20, 4032 and 8032 computers, and 1541, 2031 and 4040
disk drives (use drive zero).
0 REM 1541 USER'S GUIDE APPENDIX E
1 REM COPYRIGHT: G. NEUFELD, 1984
3 REM UNSCRATCH ANY TYPE OF SCRATCHED F
ILE
4 :
100 DIM BM-/.(35,2),TL<50),SL(50),F*(50)
110 DIM SD<50>,DE(50)
120 Z*=CHR*(0>
130 OPEN 15,8, 15
140 PRINT" {CLR} 1541 FILE UNSCRATCHER"
150 PRINT" (DOWN* INSERT DISKETTE WITH SCR
ATCHED FILE"
160 PRINT" {DOWN} PRESS CRVS}C{ROFF} TO CO
NTINUE"
170 PRINT" {RVS}Q£ROFF} TO QUIT"
180 GET A*: IF A*<>"" GOTO 180
190 GET A*: IF A*="" GOTO 190
200 IF A*="C" GOTO 230
210 IF A*="Q" THEN PRINT" <CLR}":CL0SE15:
END
220 GOTO 190
230 PRINT" CDOWN} OK. READING DIRECTORY...
400
240 PRINT#15, "IO"
250 GOSUB 1290: IF FL=1 GOTO 140
260 REM READ THE DIRECTORY
270 SE=l:NE=0
280 0PEN1,8,4,"#"
290 PRINT#15,"Ul:4 O 18"SE
300 GOSUB 1290: IF FL=1 THEN CLOSE l:G0T0
140
310 FOR K=0 TO 7
320 PR I NT# 1 5 , " B-P : 4 " 2+K*32
330 GET# 1 , A* : FT=ASC ( A*+Z * )
340 GET#1,A*:TL=ASC<A*+Z*>
350 GET#1«A*:SL=ASC<A*+Z*)
360 IF (SL=0 AND TL=0> OR FTOO GOTO 470
370 NE=NE+1
380 TL(NE)=TL
390 SL(NE)=SL
400 SD<NE)=SE
410 DE(NE)=K
420 F*<NE>=""
430 FOR L=1T015
440 GET# 1 , A* : I FA*= " " THENA*=Z*
450 F*<NE)=F*<NE)+A*
460 NEXT L
470 NEXT K
480 PRINT#15, "B-P:4"0
490 GET* 1 , A* : TL=ASC ( A*+Z* >
500 GET# 1 , A* : SE=ASC ( A*+Z* )
510 IF TL=18 GOTO 290
520 :
530 PRINT" CCLR> 1541 FILE UNSCRATCHER"
540 IF NEOO GOTO 600
550 PRINT" CDOWN>NO SCRATCHED FILES FOUND
560 PRINT" (DOWN* PRESS {RVS> RETURN <ROFF J
TO CONTINUE"
570 GET A*: IF A*<>"" GOTO 570
580 GET A*: IF A*OCHR*<13) GOTO 580
590 CLOSE l:GOTO 140
600 PRINT" CDOWN>"NE"SCRATCHED FILES FOUN
D<DOWN>"
610 FOR K=l TO NE
620 PRINTK"<:LEFT>. "F*<K>
630 NEXT
640 PRINT" {DOWN* WHICH FILE TO UNSCRATCH
<i"NE*-i"<:left>>";
650 INPUT X*:X=INT(VAL(X*)>
660 IF X<1 OR X>NE THEN CLOSE 1:G0T0 140
670 PRINT" {DOWN* READING THE BAM PRIOR TO
CHECK "
401
680 PRINT#15,"U1:4 0 18 0"
690 PRINT#15,"B-P:4 4"
700 FOR T=l TO 35
710 GET#1,A*:REM DUMP BLK COUNT
720 GET#1,A*:BM7.(T,0)=ASC<A*+Z*):REM sec
T 0-7
730 GET#1,A*:BM7.(T, 1)=ASC(A*+Z*):REM SEC
T 8-15
740 GET#1,A*:BM7.(T,2)=ASC(A*+Z*):REM SEC
T 16-23
750 NEXT T
760 PR I NT "CHECKING IF ALL BLOCKS ARE FRE
E. . . "
770 TL=TL(X):SL=SL(X):NB=0
780 SN=2+(SL<8)+(SL<16)
790 NB=NB+1
800 SP=SL-8*SN
810 IF <2'SP AND BM%(TL,SN) )=0 GOTO 1200
820 PRINT#15, "Ul:4 0"TL;SL
830 GOSUB 1290
840 PRINT#15, "B-P:4 O"
850 GET#1 , A*" TL=ASC < A*+Z«)
860 GET#1,A*:SL=ASC(A*+Z*)
870 IF TLOO GOTO 780
880 PRINT" {DOWNJFILE IS"NB"BLOCKS LONG A
ND"
890 PR I NT "CAN BE UNSCRATCHED"
900 INPUT" CDOWN>SHALL I UNSCRATCH IT <Y/
N)"5A*
910 IF <ASC(A*)AND127)<>89 THEN CLOSE 1:
GOTO 140
920 PRINT#15, "Ul:4 O 18"SD(X)
930 GOSUB 1290
940 PRINT#15, "B-P:4"23+32*DE(X)
950 GET#i,A*:A=ASC(A*+Z*>
955 PR I NT# 1 5 , " B-P : 4 " 2+32*DE < X )
960 IF A=0 GOTO 990
970 PRINT" <DOWN> THIS IS A RELATIVE FILE"
971 PRINT" RECORD SIZE="A:FT=132:G0T010
70
990 T*="P": INPUT" <DOWN>FILE TYPE <P/S/U)
"ST*
1 OOO T*=LEFT* ( T<* , 1 )
1010 IF T*="P" THEN FT=130:G0T0 1070
1020 IF T*="S» THEN FT=129:GOTO 1070
1030 IF T*="U" THEN FT=131:GOTO 1070
1040 PRINT" {DOWN* ENTER ONLY P, S, OR U"
1050 PRINT"P=PRG S=SEQ U=USR"
1060 GOTO 990
402
1070 PRINT#1,CHR*(FT);
1080 PRINT#15, "B-P:4M31+32*DE<X)
1090 BH=INT<NB/256>
llOO BL=NB-256*BH
lllO PRINT#1,CHR*(BH);
1120 PRINT#1,CHR*<BL);
1130 PRINT#15, "U2:4 O 18"SD(X)
1140 PRINT" { DOWN? NOW VALIDATING DISKETTE
1 ISO PRINT#15, "VO"
1160 GOSUB 1290:IF FL=1 THEN CLOSElrGOTO
140
1170 CLOSE 1
1180 PRINT" CDOWN> ALL DONE!"
1190 GOTO 1220
1200 PRINT" CDOWN}CAN NOT UN-SCRATCH THIS
FILE"
1210 PRINT" I DOWN} FOUND ALLOCATED BLOCK I
N BAM. "
1220 PRINT" {DOWN} PRESS {RVS}RETURN{ROFF}
TO CONTINUE"
1230 GET A*: IF A*<>"" GOTO 1230
1240 GET A*: IF A*OCHR*(13) GOTO 1240
1250 CLOSE l:GOTO 140
1260 :
1270 REM READ DISK STATUS
1280 :
1290 INPUT#15,EN,E*?ET,ES
1300 IF EN<20 THEN FL=0: RETURN
1310 PRINT" {DOWN} {RVS}DISK ERROR {ROFF}"
1320 printen;E*;ET;ES
1330 PRINT" {DOWN}PRESS {RVS}RETURN{ROFF}
TO CONTINUE"
1340 GET A*: IF A*<>"" GOTO 1340
1350 GET A*: IF A*OCHR*<13> GOTO 1350
1360 FL=l: RETURN
*********************
9. PROGRAM: PRG HEX DUMP
*********************
This program produces a hex dump to the screen of any program file on disk. The
important features of the inner structure of BASIC (line link bytes, line numbers,
and null bytes marking the end-of-line) are highlighted in color. The decimal
equivalents of the line numbers are shown on the right of the display. Illegal bytes
are highlighted in blue. Because of the color highlights this program is most
effective when used with the Commodore 64.
Compatible with C-64, VIC-20, 4032 and 8032 computers, and 1541, 2031 and 4040
disk drives (use drive zero).
403
0 REM 1541 USER'S GUIDE APPENDIX E
1 REM COPYRIGHT: G. NEUFELD, 1984
2 :
3 REM HEX DUMP OF PROGRAM FILE WITH LIN
E LINKS, LINE
4 REM NUMBERS, ILLEGAL BYTES, AND LINE
ENDS HIGHLIGHTED
5 :
lOO REM HEX ANAL
HO DIM C(255)
120 C1=30:C2=158:C3=152:C4=144:BC=11
1 30 P0KE53280 , BC : P0KE5328 1 , BC
140 N*=CHR* ( O ) : BL*=CHR* < 1 54 ) : 0L=- 1
150 PRINTCHR*(C3)"<CLR><:D0WN> HEX DUMP
OF PROGRAM FILE"
160 INPUT" {DOWN} PGM FILE NAME";F*
170 0PEN15,8, 15, "10"
180 GOSUB650: IFF=1THENST0P
190 OPENl.8,5, "0:"+F*+",P,R"
200 g0sub650: iff=1thenst0p
210 printchr*(C3)"<:clr><:down> hex dump
OF ,,F*"<D0WNJ"
220 GET# 1 , C* : L=ASC ( C*+N* )
230 GET# 1 , C* : H=ASC < C*+N* )
240 HA=L+256*H-1:N=0
250 gosub 570: z=0: print" <rvs> "chr* (c4> 5 :
g0sub700: print" {roff> ";
260 get#i,c*:pl=asc(c*+n*>:rem pointer
270 GET#1,C*:PH=ASC(C*+N*):REM pointer h
I
280 qa=256*ph+pl-2: ifqa=-2 then ci =154
290 i fqa< >-2 and ( qa< h aor < qa-h a ) >256 ) thenm
*=mi:left>bad linkm:ci=i54
300 Z =PL : PR I NT " CRVS3 " CHR* (CD;: G0SUB700
310 IF MOO THEN PRINT" £RVS> £ROFF>";
320 Z=PH: PRINT" CRVSJ "CHR* <C1 ) ; : G0SUB700:
IFNOOTHENPRINT"{ROFFJ ";
330 IF QA=-2 THENPRINT:PRINTTAB(8>"*DOWN
>{RVS> END OF PROGRAM ":CL0SE1:CL0SE15:E
ND
340 GET#1,C*:LL=ASC<C*+N*):REM LINE# LO
350 GET#1,C*:LH=ASC(C*+N*):REM line# hi
360 Z=LL: PRINT" CRVS* "CHR* (C2) ; : G0SUB700
370 IFM*=" "THENM*="#"+MID* (STR* <256*LH+L
L),2)
380 IF 0LX256*LH+LL)THENM*=BL*+"<:< — >" +
MID* (STR* (256*LH+LL> , 2)
390 0L=(256*LH+LL>
400 IF NOO THEN PRINT" <RVS> CROFFJ";
410 Z=LH: PRINT" CRVSJ "CHR* (C2) ; : GOSUB700:
IFNOOTHENPRINT"€ROFF> "s
404
420 Q=0:FORK=HATOQA
430 get#i,c*:z=asc<c*+n*>:rem CHARACTER
440 IF Z=0 GOTO 500
450 IF Q=0 AND (Z<31 OR Z>219) GOTO 500
460 IF Q=0 AND <Z>95 AND Z<128) GOTO 500
470 IF Z=34 THEN Q=2+N0TQ
480 PRINTCHR*<C3);:G0SUB700:IFNO0THEN P
RINT" ";
490 GOTO 510
500 PR I NT " CR VS> " BL* ; : M*=BL*+ " ERROR " : GOSU
B700 : I FN< >OTHENPR I NT " CROFF* " ;
510 NEXT
520 GET#1,C*:LL=ASC<C*+N*>:REM LINE END
530 IFLL=OTHENZ=LL: PRINT" CRVS> "CHR* (C4) ;
: G0SUB700: IFN< >OTHENPRINT" {ROFFJ " ;
540 IFLL=0G0T0560
550 Z=LL: PRINT" {RVSJ "BL*; : M*=BL*+" ERROR"
: G0SUB700 : I FN< >OTHENPR I NT " CROFF J " ;
560 GOTO 260
570 XL=XL+l:lF XL=20THENG0SUB750
580 ZH=INT (HA/256) : ZL=HA-256*ZH
590 Zl=(ZHAND240)/16+48: IFZ1>57THENZ1=Z1
+7
600 Z2=<ZHAND15)+48: IFZ2>57THENZ2=Z2+7
610 Z3= < ZLAND240) / 16+48: IFZ3>57THENZ3=Z3
+7
620 Z4= (ZLAND15) +48: IFZ4>57THENZ4=Z4+7
630 PRINTCHR*(C3)": *"CHR*(Z1)CHR* (Z2)CH
R*(Z3)CHR*(Z4)" ";
640 RETURN
650 INPUT#15,E,E*,T,S
660 IF E<20 THEN F=0: RETURN
670 PRINT"tDOWN>DISK ERROR"
680 PRINT" "E;E*;T;S:CI_OSEl:CLOSE15
690 F=l: RETURN
700 ZH= <ZAND240) /16+48: IFZH>57THENZH=ZH+
7
710 ZL=<ZAND15)+48: IFZL>57THENZL=ZL+7
720 PRINTCHR*<ZH);CHR*<ZL>;
730 HA=HA+ 1 : N=N+ 1 : I FN >7THENN=0 : PR I NTCHR*
< C2 ) " €ROFF> " M* : M*= " " : G0SUB570
740 RETURN
750 PRINTCHR*(C3)"<RVS> PRESS G TO GO ON
OR Q TO QUIT {ROFF>"
760 GET A*:IFA*=""G0T0760
770 IF A*="Q"THENCl_0SEl:CL0SE15:END
780 IF A*O"G"G0T0760
790 PRINT" CUPJ
800 PR I NT " {UP> " ; : XL=0 : RETURN
405
*************************
10. PROGRAM: PROGRAM ANALYZER
*************************
This program produces a detailed line by line analysis of any program file on disk.
Each byte of the program is analyzed. All the line links, line numbers, BASIC
tokens, ASCII characters, and null bytes marking the end-of-line are displayed
and interpreted. A useful program for studying the inner structure of any BASIC
program. Assumes at least a 40 column display.
Compatible with C-64, VIC-20, 4032 and 8032 computers, and 1541, 2031 and 4040
disk drives (use drive zero).
0 REM 1541 USER'S GUIDE APPENDIX E
1 REM COPYRIGHT: G. NEUFELD, 1984
2 :
3 REM BYTE BY BYTE ANALYSIS OF BASIC PR
OGRAM FILE
4 :
100 DIM A* (2, 255)
HO PRINT" tCLR>CDOWN> "TAB < 10) "PROGRAM FI
LE SCAN"
120 PRINT" {DOWN} EXCUSE ME WHILE I SET UP
SOME TABLES"
130 REM SET UP ASCII TABLE (WITHIN QUOTE
S)
140 RESTORE
150 N*=CHR*(0):D*=" {HOME? {DOWN 23>"
1 60 FORK=OT04 : A* ( 1 , K ) =CHR* ( K ) : NEXT
1 70 F0RK=5T031 : RE AD A* ( 1 , K) : NEXT
180 FORK=32T0132: A* ( 1 , K) =CHR* <K) : NEXT: A*
(1,129)="0RANGE"
1 90 FORK= 1 33T0 1 60 : READA* < 1 , K ) : NEX T
20O F0RK=161T0255: A* ( 1 , K) =CHR* <K) : NEXT
210 :
220 REM SET UP BASIC BYTES & KEYWORDS TA
BLE
230 F0RK=1T033:A*<0,K)="?SYNTAX ERROR" :N
EXT: A* (O, 32)=" "
240 F0RK=34T095: A* (O, K) =CHR* <K) : NEXT
250 F0RK=96T0127:A*(0,K)="?SYNTAX ERROR"
:NEXT
260 FOR K=128T0218: READA* (0,K): NEXT
270 F0RK=219T0255:A*<0,K>="?SYNTAX ERROR
":NEXT
280 :
290 REM TABLE OF SPECIAL ASCII CHARACTER
S
300 DATA WHITE,,, DISABLE, ENABLE,,,, RETUR
N, LOWER CASE,,
310 DATA DOWN,RVS ON, HOME, DEL, ,,,,,, ,RED
, RIGHT, GREEN, BLUE
406
320 DATA FUNC#1,FUNC#3,FUNC#5,FUNC#7,FUN
C#2 , FUNC#4 , FUNC#6 , FUNC#8
330 DATA SHIFTED RETURN, UPPER CASE, , BLAC
K,UP,RVS 0FF,CLR SCREEN, INSERT
340 DATA BROWN, LT RED,GREY1,GREY2,LT SRE
EN,LT BLUE,GREY3
350 DATA PURPLE. LEFT, YELLOW, CYAN, SHIFT S
PC
360 :
370 REM TABLE OF BASIC KEYWORDS
380 DATA END, FOR, NEXT, DATA, INPUT*, INPUT,
DIM, READ . LET , GOTO , RUN , I F , RESTORE
390 DATA 60SUB, RETURN, REM, STOP, ON, WAIT, L
OAD, SAVE, VERIFY, DEF, POKE, PRINT#
400 DATA PRINT,CONT,LIST,CLR,CMD,SYS,OPE
N, CLOSE, GET, NEW, TAB ( , TO, FN, SPC <
410 DATA THEN, NOT, STEP, + ,-,*, /,"\ AND, OR,
>, =, < , SGN, INT, ABS, USR, FRE, POS
420 DATA SQR,RND,LOG,EXP,COS,SIN,TAN,ATN
, PEEK, LEN, STR*, VAL, ASC, CHR*, LEFT*
430 DATA RIGHT*, MID*, GO, CONCAT,DOPEN,DCL
OSE, RECORD, HEADER, COLLECT, BACKUP
440 DATA COP Y, APPEND, DS A VE,DLO AD, CATALOG
, RENAME, SCRATCH, DIRECTORY
450 :
460 PRINT" CCLRXDOWN* "TAB (10) "PROGRAM FI
LE SCAN"
470 INPUT "{DOWN* PGM FILE NAME"jF*
480 OPEN 15,8,15
490 GOSUB 1190: IF DF=1 THEN STOP: REM DIS
K ERROR STATUS
500 OPEN 1,8,5, ,,0:"+F*+",P,R"
510 GOSUB 1190: IF DF=1 THEN STOPrREM DIS
K ERROR STATUS
520 GET# 1 , P* : LL=ASC ( P*+N* )
530 GE T# 1 , P* : LH=ASC ( P*+N* )
540 PRINT" CDOWN>THE FIRST TWO BYTES IN T
HE FILE ARE:"
550 Z=LL:G0SUB1240:PRINTTAB(8) "CDOWN>*"Z
*;
560 2=LH: GOSUB 1240: PR I NT" AND *"Z*
570 LD=LL+256*LH
580 PR I NT "{DOWN!* THE LOAD ADDRESS IS: "LD"
(*";
590 Z=LH: G0SUB1240: PRINTZ*; : Z=LL: G0SUB12
40:PRINTZ*">"
600 IFLD=1025THEN PRINT" {DOWN* PROGRAM WR
ITTEN ON A PET":G0T0650
610 IFLD=2049THENPRINT"CD0WN>PR0GRAM WRI
TTEN ON A 64":G0T0650
407
620 IFLD=2049THENPRINT"{D0WN}PR0GRAM WRI
TTEN ON A VIC-20":G0T0650
630 IFLD=4609THENPRINT"{D0WN}PR0GRAM WRI
TTEN ON AN EXPANDED VIC-20":G0T0650
640 PRINT" {DOWN} THIS IS NOT A BASIC PROG
RAM " : CLOSE 1 : CLOSE 1 5 : END
650 M*=" FIRST LINE": GOSUB 1270
660 SA=LD+4:AD=LD
670 L=0 : GET# 1 , P* : PL=ASC ( P*+N* )
680 GET# 1 , P* : PH=ASC ( P*+N* >
690 QA=PH*256+PL-1
700 PRINT" {CLR} PREAMBLE TO NEXT LINE"
710 PRINT" tCOM-U 22} "
720 PRINT" {DOWN} NEXT LINE POINTER BYTES:
730 PRINT" CDOWN} {RVS} ADDRESS V
ALUE "
740 Z=AD:GOSUB1320:PRINTAD;Z*5 :ad=ad+i
750 Z=PL:G0SUB1240:PRINTRIGHT*<" "+S
TR*(PL) ,5) " <*"Z*">"
760 Z=AD: GOSUB1320: PRINTAD; Z*; : AD=AD+1
770 Z=PH:G0SUB1240:PRINTRIGHT*(" "+S
TR*(PH),5)" (*"Z*">"
780 PRINT" CDOWN}NEXT LINE STARTS AT:"
790 PRINT" {DOWN} 256 *"PH"+"PL"="256*PH
+PL" (*";
800 Z=PH : GOSUB 1 240 : PR I NT Z * ; : Z =PL : GOSUB 1 2
40:PRINTZ*") "
810 IF PHOO OR PLOO GOTO 860
820 PRINT" {DOWN} END OF BASIC PROGRAM"
830 IFST=64THENPRINT"{D0WN}ALS0 END OF P
ROGRAM FILE":GOT0850
840 PRINT" {DOWN} BUT, NOT THE END OF THE
PGM FILE."
850 CLOSE 1: CLOSE 15: END
860 GET# 1 , P* : LL= ASC < P*+N* )
870 GET# 1 , P* : LH=ASC ( P*+N* )
880 PRINT" {DOWNLINE NUMBER BYTES:"
890 PRINT" {DOWN} {RVS} ADDRESS V
ALUE "
900 Z= AD : GOSUB 1 320 : PR I NT AD ; Z * ; : AD=AD+ 1
910 Z=LL:G0SUB1240:PRINTRIGHT*(" "+S
TR*(LL),5)" <*"Z*">"
920 Z=AD: G0SUB1320: PRINTAD; Z*; : AD=AD+1
930 Z=LH:G0SUB1240:PRINTRIGHT*<" "+S
TR*<LH),5)" (*"Z*")"
940 PRINT" {DOWN}LINE NUMBER IS: "
950 PRINT" {DOWN} 256 *"LH"+"LL"="256*LH
+LL"(*";
960 Z=LH: G0SUB1240: PRINTZ*; : Z=LL: GOSUB12
40:PRINTZ*"> "
408
970
9SO
+LL
990
UE
1000
1010
1020
1030
1040
M*="LINE ANALYSIS" : GOSUB 1270
PRINT" {CLR} ANALYSIS OF LINE #"256*LH
ADDRESS
print" {down* {rvsj
meaning {down}"
l=o:q=o
for ma=satoqa
z=ad: gosub 1320: pr i nt ad; z*j :
get#1 , p*: p=asc <p*+n*>
z=p : gosub 1 240 : pr i ntr i ght* ( "
R*<P),4)" <*"Z*">";
1050 IF P=0 THEN PRINT
VAL
AD=AD+1
'+ST
{RVS}LINE END":G
THEN PRINT" ASC- "A*(1,P)
AND P>127 THEN PRINT" TOK-
OT01100
1060 IF Q=l
1070 IF Q=0
A*<0,P)
1080 IF Q=0 AND P<128 THEN PRINT" CHR-
A*(0,P)
1090 IF
1100 L=L+l;
01150
lllO M*="MORE ANALYSIS": GOSUB 1270
PRINT" {CLR} ANALYSIS OF LINE #"256*L
P=34 THEN Q=2+N0TQ
IF L/18OINT(L/18)0RMA=QA
GOT
1120
H+LL
1130
LUE
1140
1150
1160
1170
1180
1190
1200
1210
1220
1230
1240
+7
1250
1260
1270
QUIT"
1280
1290
D
1300
1310
1320
ADDRESS
VA
PRINT" {DOWN} <RVS}
MEANING {DOWN}"
L=0
NEXT
SA=AD+4:M*="NEXT LINE":G0SUB1270
G0T0670
NEXT: STOP
INPUT#15,E,E*,T,S
IF E<20 THEN DF=0: RETURN
PRINT" {DOWN} DISK ERROR"
PRINT" "E;e*;t;s
DF=l: RETURN
ZH=48+(ZAND240> /16: IFZH>57THENZH=ZH
ZL=48+(ZAND15) : IFZL>57THENZL=ZL+7
Z*=CHR* ( ZH> +CHR* ( ZL) : RETURN
PRINTD*; "PRESS C FOR "M*; " OR Q TO
GET X * : I F X *= " Q " THENCLOSE 1 : CLOSE 1 5 : EN
IFX*<>"C"GOT01290
RETURN
ZH=INT < Z/256) : Z 1=48+ ( ZHAND240) / 16: I
FZl>57THENZl=Zl+7
1330 Z2=48+(ZHAND15) : IFZ2>57THENZ2=Z2+7
409
1340 ZL=Z-256*ZH: Z3=48+<ZLAND240> /16: IFZ
3>57THENZ3=Z3+7
1350 Z4=48+(ZLAND15) : IFZ4>57THENZ4=Z4+7
1360 Z*=" (*"+CHR* (Z 1 ) +CHR* <Z2> +CHR* < Z3> +
CHR*(Z4)+") "
1370 RETURN
410
INDEX
accessing a file 92-93
ASCII text files 163-164
asterisk (*) 21-22
at sign (@) 38, 42
auto-run 25
B
BAM ..
BASIC .
..76
.124
housekeeper program .
79-87
loading with DOS 5.1 39
saving with DOS 5.1 41-42
token 126-127
binary chop 205
alternate uses for 209-210
BLOCK- ALLOCATE 331, 357-358
Block Availability Map 76
BLOCK-EXECUTE 358
BLOCK-FREE 358
BLOCK-READ 330, 356-357
BLOCK-WRITE 330, 357
BUFFER-POINTER 356
bugs in DOS 2.6 330-331
bump 278
bytes in a BASIC program 125-127
changing the device number 276-278
channel number 93, 96-97
CHKIN 228
CHKOUT
CHRIN ...
.230
.228
CHROUT
CLALL
cleaning the record/play head .
CLOSE
closing a file
CLRCHN
230
231
276
. 53-55, 231-232
110-111
combining files
command channel
closing
opening
. 229, 230-231
73-76
47,96-97
53-55
47-48
reading 50-51
commands, DOS 5.1 38
at sign (@) 38, 42
greater then symbol (>) 38, 42
left arrow (— ) 41-42
percent sign (%) 39-41
slash (/) 39-40
up arrow (t) 41
Commodore 64, rear view 16
concatenate 355
COPY
combining files
crossed files
. 71-73, 355
73-76
76
D
data 89
data block 61, 297, 299
identifier 299
data bytes 299
data checksum 299
data file 90-91
delimiter 90-91
device number 93, 95-96
jumpers 277
directory 19-21, 134-137
display using DOS 5.1 38-39
organization 299
selective listings 22-23
disk commands 48-49
disk drive 15
alignment 278
long board 276, 277
rear view 16
short board 276, 277
top view with circuit board removed 286
disk drive status 101-103, 234-235
diskette housekeeping 57
BASIC housekeeper program 79-87
copying a file 71-73
formatting 58-61
initializing a diskette 64-66
renaming a file 68-71
reusing a diskette 61-64
scratching a file 66-68
validating a diskette 76-79
disk operating system (DOS) 42
DOS 5.1 37
housekeeping commands 42
loading 37-38
loading and running a program 41
saving a program 41-42
using commands in programs 43
wild cards 39, 41
drive wheel slippage, prevention 287-289
E
end-of-file value 109
error status 42-43
reading 51
in immediate mode 52-53
using GET* 51-52
using INPUT* 51-52
411
F
field,
file ..
.90-91
91
accessing 92-93
closing 110-111
identification information 93, 97-101
locked 92
number 93, 94-95
opening 93-94
program 91-92
reading from 106
relative 91-92
sequential 91-92
unclosed 92
user 91-92
writing to 103-105
file handling 89
FL 209
flag variable 209
flippy 293-294
floppy diskette 17, 291-292
taking care of 18-19
forbidden record lengths 178
formatting a diskette 29-30, 58-61
forward pointers 313
G
gap .
GCR.
..61
.295
general operating hints 275-276
GET# 50-51
reading the error status 51-52
greater than symbol (>) 38, 42
Group Code Recording 295
H
header
header block
identifier
header checksum
header gap
..61
.297
.298
.297
housekeeping commands, DOS 5.1 42
hysteresis 281
immediate mode 52-53
indexed relative files 201
updating the index 204-205
index files 174
INITIALIZE 64-66, 356
using DOS 5.1 64
using the command channel 65
initializing a diskette 64-66
INPUT* 50-51
51-52
17
180
180
232
232-234
reading the error status .
inserting a diskette
INT
integer
I/O status
using READST .
L
left arrow («-) 41-42
LIST 25, 28
listing programs on a printer 28-29
LOAD 24
load address 114
loading programs 24-28
DOS 5.1 37-38
locked file 92
logical file number 93, 94-95
M
machine language file handling 225
closing a file 231
using CLALL 231
using CLOSE 231-232
disk commands 235
opening a file 225
using OPEN 226-228
using SETLFS 226
using SETNAM 225-226
reading a file 228
using CHKIN 228
using CHRIN 228
using CLRCHN 229
writing a file 229
using CHKOUT 230
using CHROUT 230
using CLRCHN 230-231
machine language program 26
verifying 34
MEMORY-EXECUTE 359
MEMORY-READ 358-359
MEMORY-WRITE 359
modifying the directory 307
monitoring the disk drive status 234-235
mouse bites 292
mylar disk 292
IM
NEW 29-30, 58, 60-61, 353
short NEW 61-64, 354
using DOS 5.1 59
using the command channel 60
newing a diskette 29-30, 58
number of blocks free 105
o
OPEN ...
.......... .„„„u„aJI_,
.. 47-48, 49, 93-94
opening a
file..
93-94
opening the command channel
... 47-48
jumpers, device number .
.277
P
pattern matching 21-22
percent sign (%) 39-41
POSITION 179, 180
powering on 16-17
PRINT* 48-49, 103-104
program file 91-92, 113
file storage 319-320
loading 115
fnmi immediate mode 115-116
fpom within a program 116-118
opening 130-132
412
. 133-134
123
. 123-124
reading
replacing
from immediate mode
from within a program 124
replacing an existing file 132
saving 118
from immediate mode 118-120
from within a program 120-121
verifying 121
122-123
123
132-133
from immediate mode ..
from within a program
writing
question mark (?)
21-22
R
reading from a file 106
reading the command channel 50-51
read/write head 294
record 90-91
fixed length fields 174-175, 176
forbidden lengths 178
variable length fields 175, 176
recording gap 294
recovering a damaged directory 260-266
recovering a damaged file 267-269
recovering a physically damaged diskette 272
recovering data from an unclosed file 254-260
recovering from a bad replacement 269-272
recovering from a full NEW 266-267
recovering from a short NEW 266-267
relative files 173
adding to 192
closing 185-186
file storage 320
main data file 320-325
opening 177-179
positioning to a record 180-182
reading 183-185
side-sector file 325-328
writing a record 182-183
relocated program 34
RENAME 68-71, 355-356
using DOS 5.1 69
using the command channel 69-70
renaming a file 68-71
reusing a diskette 61-64
S
SAVE 31, 32
saving a program 30-33
SCRATCH 66-68, 354-355
using DOS 5.1 67
using the command channel 67-68
scratching a file 66-68
secondary address 93, 96-97
sector 60-61
number 298
selective directory listings 22-23
self-modifying programs 137
sequential files 147
appending data 158-160
closing 166
file storage 318-319
opening 148-151
reading . 152-156
replacing 156-158
writing 151-152
sequential program files 164-165
SETLFS 226
SETNAM 225-226
setting up the system 15-16
short NEW 61-64
using DOS 5.1 62
using the command channel 62-63
6502 microprocessor 328
file manager 328
floppy disk controller 329
slash (/) 39-40
SPACE BAR 39
ST 109-110
status variable 109-110
SUPER BAM 251-252
T
track.
number
track and sector arrangement ...
60-61
298
60
U
unclosed file 254
unscratching a file 252-254
up arrow (t) 41
user files 91-92
file storage 320
using a 1541 with a VIC-20 18
V
VALIDATE 77-79, 354
using DOS 5.1 77
using the command channel 78
validating a diskette 76-79
VERIFY 33
verifying a program 33-35
relocated program 34
machine language routine 34
VIC-20 18
loading DOS 5.1 37
viewing a directory 19-21
W
wedge
wild cards .
37
. 21-22
with DOS 5.1
. 39, 41
write protect notch 292
writing to a file 103-105
413
1541 User's
All the programs listed in 15kl User's Guide on
one diskette! Save yourself hours of typing and
debugging.
Only $24.95 with this couponl
Available only to readers of this book. Send $24.95
plus $2.00 shipping and handling (California
residents add &/*% sales tax) to:
M DATAMOSTT
19821 Nordhoff Street, Northridge, CA 91324
(818) 709-1202
Please RUSH me 15hl User's Guide diskette. I have enclosed
my check or money order for $24.95 plus $2.00 shipping and
handling (California residents add $1.62 for sales tax).
Name _
Address
City
Phone (
.State.
.Zip.
MORE GREAT BOOKS FROM DATAMOST
InSlClO COnilllOdOrO DOS — Now you can get the inside story on
the 1541 disk operating system. This valuable tool for the intermediate and
advanced programmer includes complete information on correcting errors
and omissions in Commodore's 1541 User's Manual, diskette formatting, file
storage, reading and writing data in non-standard ways, backing up
protected disks, recovering damaged data, and more. Includes an analysis of
154l's ROM, completely disassembled and annotated. $19-95
The Elementary Commodore
Explains the Commodore 64 in
simple, everyday language. How
to hook it up, use the keyboard,
and program in BASIC. Teaches
about word processing, utilities
and peripherals. $14.95
Oames Commodores Play
A collection of classic computer
games that teaches BASIC using a
games, and graphics approach.
Simply type them in and make
your own modifications. $14.95
Intermediate Commodore
Takes you from being a fledging
programmer and teaches important
principles so you can handle
more complicated problems.
Helps'you take that step from
elementary BASIC programming
to machine language
programming. $14.95
C-64 Came Construction Kit
Shows you how to write your
own BASIC games! This unique
book gives examples of different
games and teaches fundemental
lessons of quality game
programming. $14.95
Computers & Writing
An innovative approach to
teaching children to write using
the computer. Topics discussed
include computers in the classroom,
adapting traditional teaching
methods to include computers,
setting up a creative environment
at home and more. For both
parents and teachers. $9.95
A Shortcut Through
Adventureland, Volume I
Contains answers to 14 of the -
most popular hi-res puzzle solving
adventures including Wizard and
the Princess, Cranston Manor,
The Dark Crystal, Escape from
Rungistan, Time Zone and
more. $9-95
The C-64 Home Companion
This is the book that should have
come with your Commodore 64.
Straight answers to home computing
questions, dozens of software
reviews, BASIC and more! $19-95
The Musical Commodore
Introduces you to music theory
and computing at the same time.
For beginners as well as pros, this
book helps you turn your C-64
into a musical instrument. $14.95
Super Computer Snooper (C-64)
Learn how the computer
"thinks." Investigate memory,
screens, programs and variables,
keyboards, printers, and
expansion boards. $14.95
Programming for Profit
Aimed at the programmer who
wants to get his or her software
published. This is a unique
guidebook through the maze of
traditions, rules and standards in
the software Industry. Contains
hints and tricks of the trade
and much more! $14.95
C-64 Logo Workbook
Teaches children in grades 2-6
how the Logo language can be
used for problem solving. Learn
| about the "turtle," variables,
I geometry and recursion. $12.95
I Kids to Kids on the C-64
I Written by kids for kids, explains
BASIC programming in simple,
straight-forward language. Two
chapters are devoted to sound
and graphics, and another shows
how to write an original game. $9.95
A Shortcut Through
Adventureland, Volume li
The cheater's guide to all of
Infocom's text adventures to
date: The Zork Trilogy, Infidel,
The Witness, Deadline and
Enchanter to name a few. $9.95
Available at better book and computer stores, or contact
H DATAMOST8
19821 Nordhoff Street, Northridge, CA 91324
(818) 709-1202
$19.95
&sx^T2*-.
tr**^'?*
's th0*toBntlal9*;H'na,in'a» cop, °*»af.
-acZ^o^r--.^^^^^
''"•WL^tfe
>-S«
"a^sa*^
""'que
^^<,~V"*
Live Music Archive
Librivox Free Audio
Metropolitan Museum
Cleveland Museum of Art
Internet Arcade
Console Living Room
Open Library
American Libraries
TV News
Understanding 9/11